Practical Recording Techniques, Fifth Edition

Practical Recording Techniques, Fifth Edition
Practical Recording Techniques
Fifth Edition
This page intentionally left blank
Practical Recording Techniques
The Step-by-Step Approach to Professional
Audio Recording
Fifth Edition
Bruce Bartlett
Jenny Bartlett
AMSTERDAM • BOSTON • HEIDELBERG • LONDON • NEW YORK • OXFORD
PARIS • SAN DIEGO • SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO
Focal Press is an imprint of Elsevier
Focal Press is an imprint of Elsevier
30 Corporate Drive, Suite 400, Burlington, MA 01803, USA
Linacre House, Jordan Hill, Oxford OX2 8DP, UK
Copyright © 2009, by Elsevier Inc. All rights reserved.
No part of this publication may be reproduced, stored in a retrieval system, or
transmitted in any form or by any means, electronic, mechanical, photocopying,
recording, or otherwise, without the prior written permission of the publisher.
Permissions may be sought directly from Elsevier’s Science & Technology Rights
Department in Oxford, UK: phone: (⫹44) 1865 843830, fax: (⫹44) 1865 853333,
E-mail: permissions@elsevier.com. You may also complete your request on-line
via the Elsevier homepage (http://elsevier.com), by selecting “Support & Contact”
then “Copyright and Permission” and then “Obtaining Permissions.”
Library of Congress Cataloging-in-Publication Data
Bartlett, Bruce.
Practical recording techniques / Bruce Bartlett, Jenny Bartlett. — 5th ed.
p. cm.
Includes index.
ISBN 978-0-240-81144-4 (pbk. : alk. paper) 1. Magnetic recorders and
recording—Handbooks, manuals, etc. 2. Sound—Recording and
reproducing—Digital techniques—Handbooks, manuals, etc.
3. Sound—Recording and reproducing—Handbooks, manuals, etc.
I. Bartlett, Jenny. II. Title.
TK7881.6.B367 2008
621.389’3–dc22
2008030765
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library.
ISBN: 978-0-240-81144-4
For information on all Focal Press publications
visit our website at www.books.elsevier.com
08
09
10
11
5
4
3
2
1
Printed in the United States of America
To family, friends, and music.
This page intentionally left blank
Contents
DEDICATION ........................................................................................................................................v
PREFACE .............................................................................................................................................ix
ACKNOWLEDGMENTS ..................................................................................................................xiii
CHAPTER 1
CHAPTER 2
CHAPTER 3
CHAPTER 4
CHAPTER 5
CHAPTER 6
CHAPTER 7
CHAPTER 8
CHAPTER 9
CHAPTER 10
CHAPTER 11
CHAPTER 12
CHAPTER 13
CHAPTER 14
CHAPTER 15
CHAPTER 16
CHAPTER 17
CHAPTER 18
CHAPTER 19
CHAPTER 20
APPENDIX A
APPENDIX B
APPENDIX C
APPENDIX D
Music: Why You Record .................................................................................... 1
The Recording Chain .........................................................................................5
Sound, Signals, and Studio Acoustics .........................................................15
Equipping Your Studio ................................................................................... 37
Monitoring .........................................................................................................67
Microphones .....................................................................................................79
Microphone Technique Basics ................................................................... 101
Microphone Techniques...............................................................................123
Digital Recording ...........................................................................................165
Effects and Signal Processors ...................................................................195
Mixers and Mixing Consoles ...................................................................... 237
Mixer Operation ............................................................................................ 257
Computer Recording....................................................................................289
Judging Sound Quality ................................................................................. 331
Session Procedures, Mastering, and CD Burning .................................357
The MIDI Studio: Equipment and Recording Procedures ...................381
On-Location Recording of Popular Music..............................................407
On-Location Recording of Classical Music ............................................439
Surround Sound: Techniques and Media................................................ 457
Web Audio and Online Collaboration.......................................................493
dB or Not dB ..................................................................................................509
Optimizing Your Computer for Multitrack Recording...........................521
Impedance ...................................................................................................... 535
Where to Learn More ...................................................................................541
GLOSSARY ..................................................................................................................................... 547
INDEX ...............................................................................................................................................607
Practical Recording Techniques Demo Web Site ...................................................................631
ONLINE at www.elsevierdirect.com/companions/9780240811444
See the Practical Recording Techniques Demo Web Site section for the
passcode to access this site.
vii
This page intentionally left blank
Preface
Recording is a highly skilled craft combining art and science. It
requires technical knowledge as well as musical understanding and
critical listening ability. By learning these skills, you can capture a
musical performance and reproduce it with quality sound for the
enjoyment and inspiration of others.
Your recordings will become carefully tailored creations of which
you can be proud. They will be a legacy that can bring pleasure to
many people for years to come.
This book is intended as a hands-on, practical guide for beginning
recording engineers, producers, musicians—anyone who wants to
make better music recordings by understanding recording equipment
and techniques. I hope to prepare the reader for work in a home studio, a small professional studio, or an on-location recording session.
Practical Recording Techniques offers up-to-date information on the
latest music recording technology, such as hard-disk and Flash memory recorders, computer recording, loop-based recording, keyboard
and digital workstations, MIDI, surround sound, Web audio, and
online collaboration. But it also guides the beginner through the
basics, showing how to make quality recordings with the new breed
of inexpensive home-studio equipment.
The first chapter answers the question, “Why do we record?” Next, the
book overviews the recording chain to instill a system concept. The
basics of sound, signals and studio acoustics are explained so that
you’ll know what you’re controlling when you adjust the controls
on a piece of recording equipment or build a studio. Then advice is
given on equipping a home studio for any style of recording, with
examples of equipment manufacturers.
Studio setup is covered next, including cables and connections,
choosing monitor speakers, and preventing hum.
ix
x
PREFACE
Each piece of recording equipment is explained in detail, as well as
the control-room techniques you’ll use during actual sessions. Two
chapters are devoted to the technology of digital recording, audio for
video and MIDI sequencing. A major chapter on computer recording
covers the latest ways of creating and recording music. Two sections
on remote recording cover techniques for both popular and classical
music.
A special chapter explains how to judge recordings and improve
them. The engineer must know not only how to use the equipment,
but also how to tell good sound from bad.
The latest developments in recording are surround sound, Web
audio, and online collaboration. All these topics are covered in detail
in their own chapters.
Finally, four appendices explain the decibel, suggest how to optimize
your computer for multitrack recording, explain impedance, and suggest further education.
On the book’s Web site at www.elsevierdirect.com/companions/
9780240811444 are audio samples that demonstrate various topics
explained in the book. Throughout the text, references to specific
Web site tracks guide the reader to relevant audio demonstrations.
Based on my work as a professional recording engineer, this book
is full of tips and shortcuts for making great-sounding recordings,
whether in a professional studio, project studio, on-location, or at
home. You’ll find many topics not covered in similar texts:
Loop-based recording
Hum prevention tips
The latest monitoring methods
Examples of mic models by type
Microphone selection guide
Tonal effects of microphone placement
Glossary of sound-quality descriptions
The latest types of digital recorders
Up-to-date coverage of computer recording
Optimizing your computer for multitrack recording
Documenting the recording session
Audio-for-video techniques
On-location recording
PREFACE
Troubleshooting bad sound; guidelines for good sound
Audio on the Web
Selling your music online
File sharing for online collaboration
Surround sound, DVD, and Blu-ray
xi
This page intentionally left blank
Acknowledgments
Thank you to Nick Batzdorf of Recording magazine for giving me permission to draw from my “Take One” series. For my education, thank
you to the College of Wooster, Crown International, Shure Brothers
Inc., Astatic Corporation, and all the studios I’ve worked for.
Thank you to Catharine Steers, Beth Howard, Emma Baxter, Becky
Golden-Harrow, Dawnmarie Simpson, and Joanne Tracy at Focal
Press/Elsevier for their fine work and support.
My deepest thanks to Jenny Bartlett for her many helpful suggestions
as a layperson consultant and editor. She made sure the book could
be understood by beginners.
A note of appreciation goes to Sting, Uncle Earl, The Beatles, Led
Zepplin, the Pat Metheny Group, and Samuel Barber (“Adagio for
Strings”), among many others, whose music inspired the chapter
“Music: Why You Record.”
We appreciate the following manufacturers who provided photos of
their products: Edirol, Boss, Korg, Alesis, Mackie, Frontier Design,
Crown International, Audio-Technica, M-Audio, TASCAM, Zoom,
Eventide, Rane, Lexicon, CamelAudio, Izotope, Line 6, RML Labs,
E-MU, PreSonus, Digidesign, Cakewalk, Propellerhead Software,
CD Baby, TuneCore, and imeem (Snocap, Inc.).
Finally, to the musicians I’ve recorded and played with, a special
thanks for teaching me indirectly about recording.
xiii
This page intentionally left blank
CHAPTER 1
Music: Why You Record
As you learn about recording techniques for music, it’s wise to
remember that music is a wonderful reason for recording.
Music can be exalting, exciting, soothing, sensuous, and fulfilling. It’s
marvelous that recordings can preserve it. As a recording engineer or
recording musician, it’s to your advantage to better understand what
music is all about.
Music starts as musical ideas or feelings in the mind and heart of
its composer. Musical instruments are used to translate these ideas
and feelings into sound waves. Somehow, the emotion contained in
the music—the message—is coded in the vibrations of air molecules.
Those sounds are converted to electricity and stored magnetically
or optically. The composer’s message manages to survive the trip
through the mixing console and recorders; the signal is transferred
to disc or computer files. Finally, the original sound waves are reproduced in the listening room, and miraculously the original emotion
is reproduced in the listener as well.
Of course, not everyone reacts to a piece of music the same way, so
the listener may not perceive the composer’s intent. Still, it’s amazing
that anything as intangible as a thought or feeling can be conveyed by
tiny magnetic patterns on a hard disk or by pits on a compact disc.
The point of music lies in what it’s doing now, in the present. In
other words, the meaning of “Doo wop she bop” is “Doo wop she
bop.” The meaning of an Am7 chord followed by an Fmaj7 chord is
the experience of Am7 followed by Fmaj7.
1
2
Chapter 1 Music: Why You Record
INCREASING YOUR INVOLVEMENT IN MUSIC
Sometimes, to get involved in music, you must relax enough to
lie back and listen. You have to feel unhurried, to be content to sit
between your stereo speakers or wear headphones, and listen with
undivided attention. Actively analyze or feel what the musicians are
playing.
Music affects people much more when they are already feeling the
emotion expressed in the song. For example, hearing a fast Irish reel
when you’re in a party mood, or hearing a piece by Debussy when
you’re feeling sensuous, is more moving because your feelings resonate with those in the music. When you’re falling in love, any music
that is meaningful to you is enhanced.
If you identify strongly with a particular song, that tells you something about yourself and your current mood. And the songs that
other people identify with tell you something about them. You can
understand individuals better by listening to their favorite music.
DIFFERENT WAYS OF LISTENING
There are so many levels on which to listen to music—so many ways
to focus attention. Try this. Play one of your favorite records several
times while listening for these different aspects:
■
■
■
■
■
■
■
■
■
■
Overall mood and rhythm
Lyrics
Vocal technique
Bass line
Drum fills
Sound quality
Technical proficiency of musicians
Musical arrangement or structure
Reaction of one musician to another musician’s playing
Surprises versus predictable patterns
By listening to a piece of music from several perspectives, you’ll get
much more out of it than if you just hear it as background. There’s
a lot going on in any song that usually goes unnoticed. Sometimes
when you play an old familiar record and listen to the lyrics for the
first time, the whole meaning of the song changes for you.
Music: Why You Record Chapter 1
Most people react to music on the basic level of mood and rhythmic motivation. But as a recording enthusiast, you hear much more
detail because your focus demands sustained critical listening. The
same is true of trained musicians focusing on the musical aspects of
a performance.
It’s all there for anyone to hear, but you must train yourself to hear
selectively and to focus attention on a particular level of the multidimensional musical event. For example, instead of just feeling excited
while listening to an impressive lead-guitar solo, listen to what the
guitarist is actually playing. You may hear some amazing things.
Here’s one secret of really involving yourself in recorded music:
Imagine yourself playing it! For example, if you’re a bass player, listen to the bass line in a particular record, and imagine that you’re
playing the bass line. You’ll hear the part as never before. Or respond
to the music visually—see it as you do in the movie Fantasia.
Follow the melody line and see its shape. Hear where it reaches up,
strains, and then relaxes. Hear how one note leads into the next.
How does the musical expression change from moment to moment?
There are times when you can almost touch music: some music has
a prickly texture (many transients, emphasized high frequencies);
some music is soft and sinuous (sine-wave synthesizer notes, soaring vocal harmonies); and some music is airy and spacious (lots of
reverberation).
WHY RECORD?
Recording is a real service. Without it, people would be exposed to
much less music. They would be limited to the occasional live concert or to their own live music, played once and forever gone.
With recordings, you can preserve a performance for thousands
of listeners. You can hear an enormous variety of musical expressions
whenever you want. Unlike a live concert, a record can be played
over and over for analysis. Tapes or discs are also a way to achieve
a sort of immortality. The Beatles may be gone, but their music
lives on.
Records can even reveal your evolving consciousness as you grow
and change. A computer audio file or CD stays the same physically,
3
4
Chapter 1 Music: Why You Record
but you hear it differently over the years as your perception changes.
Recordings are a constant against which you can measure change in
yourself.
Be proud that you’re contributing to the recording art—it is done in
the service of music.
CHAPTER 2
The Recording Chain
Welcome to the brave new world of 21st-century recording! The
digital technologies of the past few years have given us possibilities
undreamed of just 10 years ago. This book will show you an overview of current recording technology, help you choose the equipment that best suits your needs, and guide you in using it to create
great recordings. And it will explain the technical jargon in plain
English.
Thanks to the shift from analog to digital technology, the excitement
and satisfaction of recording are accessible to more people than ever
before. It used to take a whole roomful—or truckful—of expensive
equipment to produce a good recording. But the new generation of
smaller, cheaper gear means you may be able to tuck your studio
into a corner of your bedroom or the back seat of your Toyota. As a
result, many more people are involved in the process of recording—
as musicians recording their own albums, or as engineers offering
services to others.
As a recording engineer, you’re a key player. Your skills help artists
realize their visions in sound. Your miking techniques capture the
vibrancy of the performance, whether it’s the shimmering overtones
of a string quartet or the sonic assault of an electric blues band. Your
“post” work in the studio—adding effects, tweaking levels, etc.—
will take the raw material of the performance and shape and blend
it into a polished musical statement. By mastering the technology
and becoming fluent with the audio tools at hand, you will produce
exciting recordings that will delight your clients and give you a real
sense of pride and achievement.
5
6
Chapter 2 The Recording Chain
Be sure to practice what you learn in this book. There’s no substitute
for hands-on experience. You might offer to record a band’s rehearsal
for free while you experiment and master the gear. Be patient, let
yourself make mistakes, and above all, listen to how the sound
changes when you move a mic or tweak a knob.
TYPES OF RECORDING
Let’s get started. Currently there are six main ways to record music:
1. Live stereo recording
2. Live mix recording
3. Multitrack recorder and mixer
4. Stand-alone Digital Audio Workstation (DAW; recorder–mixer)
5. Computer DAW
6. MIDI sequencing
Live Stereo Recording
Record with a stereo microphone or two microphones into a
recorder. This method is most common when used to record an
orchestra, symphonic band, pipe organ, small ensemble, quartet,
or soloist. The microphones pick up the overall sound of the instruments and the concert-hall acoustics from several feet away. You
might use this minimalist technique to record a folk group, rock
group, or acoustic jazz group in a good-sounding room.
Figure 2.1 shows the stages of this method—the links in the recording chain. From left to right,
1. The musical instruments or voices make sound waves.
2. The sound waves travel through the air and bounce or reflect
off the walls, ceiling, and floor of the concert hall. These reflections add a pleasing sense of spaciousness.
3. The sound waves from the instruments and the room reach the
microphones, which convert the sound into electrical signals.
4. The sound quality is greatly affected by mic technique: microphone choice and placement.
5. The signals from the microphones go to a 2-track recorder. It
may be a hard-drive recorder, CD-R burner, DVD-R burner, Flash
memory recorder, or computer hard drive. The signal changes
to a pattern stored on a medium, such as magnetic patterns on
The Recording Chain Chapter 2
INSTRUMENTS AND
ROOM ACOUSTICS
2-TRACK
RECORDER
MICS
HEADPHONES
Signals
MIC TECHNIQUE
FIGURE 2.1
The recording chain for live stereo recording.
a hard disk. During playback, the patterns on the medium are
converted back into a signal.
During recording, signals are stored along a track—a path or
channel on the medium containing a recorded signal. A single
medium can record one or more tracks. For example, a 2-track
hard-disk recording stores two tracks on hard disk, such as the
two different audio signals required for stereo recording.
6. To hear the signal you’re recording, you need a monitor system: headphones or a stereo power amplifier and loudspeakers.
You use the monitors to judge how well your mic technique is
working.
The speakers or headphones convert the signal back into sound. This
sound resembles that of the original instruments. Also, the acoustics
of the listening room affect the sound reaching the listener.
Live Mix Recording
This method is seldom used except for live broadcasts or recordings
of PA mixes. Using a mixer, you set up a mix of several microphones
and record the mixer’s output signal on a 2-track recorder (CD-R,
Flash memory recorder, or computer hard drive). Each mic is close
to its sound source. Figure 2.2 shows this method.
Multitrack Recorder and Mixer
Record with several mics into a mixer, which is connected to a multitrack hard-disk recorder. You record the signal of each microphone on
its own track, then mix these recorded signals after the performance
7
8
Chapter 2 The Recording Chain
INSTRUMENTS
MIXER
2-TRACK
RECORDER
MICS
HEADPHONES
OR SPEAKERS
MIC TECHNIQUE
FIGURE 2.2
The recording chain for live-mix recording.
INSTRUMENTS
MIXER
MULTITRACK
RECORDER
MICS
MIC TECHNIQUE
OR
RECORDER–MIXER
FIGURE 2.3
The recording chain for multitrack recording.
is done. You can also record different groups of instruments on each
track. Figure 2.3 shows the stages in this method.
1. Place microphones near the instruments.
2. Plug the mics into a mixing console: a big, sophisticated mixer.
During multitrack recording, the mixing console amplifies
the weak microphone signals up to the level needed by the
recorder. The console also sends each microphone signal to the
desired track.
3. Record the amplified mic signals on the multitrack recorder.
You can record more instruments later on unused tracks—a process
called overdubbing. Wearing headphones, the performer listens to
The Recording Chain Chapter 2
MIXER
MULTITRACK
RECORDER
2-TRACK
RECORDER
OR
EFFECTS
RECORDER–MIXER
FIGURE 2.4
The recording chain for a multitrack mixdown.
the recorded tracks and plays or sings along with them. You record
the performance on an unused track.
After the recording is done, you play all the tracks through the mixing console to mix them with a pleasing balance (Figure 2.4). Here
are the steps:
1. Play back the multitrack recording of the song several times,
adjusting the track volumes and tone controls until the mix is
just the way you want it. You can add effects to enhance the
sound quality. Some examples are echo, reverberation, and
compression (explained in Chapter 10). Effects are made by
signal processors that connect to your mixer, or by software
applications (plug-ins) that are part of a recording program.
2. Record or export your final stereo mix onto your computer
hard drive or to an external 2-track recorder such as the Alesis
Masterlink.
Stand-alone DAW (Recorder–Mixer)
This is a multitrack recorder and a mixer combined in one portable
chassis. It’s relatively easy to use. The recording medium is a hard
drive or a Flash memory card. Other names for a recorder–mixer are
“stand-alone Digital Audio Workstation,” “digital multitracker,” “personal digital studio,” or “portable studio.” Most recorder–mixers have
built-in effects. Figure 2.5 shows an example of a recorder–mixer.
9
10
Chapter 2 The Recording Chain
FIGURE 2.5
TASCAM DP-02, an example of a recorder–mixer.
MONITOR
HARD
DRIVE
PCI SLOT, USB PORT,
OR FIREWIRE PORT
COMPUTER
MOUSE
KEYBOARD
ALESIS
FIREPORT
AUDIO INTERFACE
SOUND CARD
CONTROLLER OR MIXER
WITH I/O
AUDIO INTERFACES (CHOOSE ONE)
FIGURE 2.6
Computer with a choice of audio interface and recording/editing software.
Computer DAW
This low-cost system includes a computer, recording software, and a
sound card or audio interface that gets audio into and out of your
computer (Figure 2.6). You record on the computer’s hard drive.
The Recording Chain Chapter 2
Using the recording software, you perform these operations:
1. Record music on the computer’s hard drive.
2. Edit the tracks to fix mistakes, delete unwanted material, or to
copy/move song sections.
3. Mix the tracks with a mouse or controller by adjusting virtual
controls that appear on your computer screen.
You might also assemble a song from samples or from loops.
Samples are recordings of single notes of various instruments. Loops
are repeating musical patterns.
MIDI Sequencing
With this recording method, a musician performs on a MIDI controller, such as a piano-style keyboard or drum pads. The controller puts
out a MIDI signal, a series of numbers that indicates which keys were
pressed and when they were pressed. The MIDI signal is recorded
into computer memory by a sequencer or sequencer program in a
computer. When you play back the MIDI sequence, it plays the tone
generators in a synthesizer or sound module. The synthesizer can be
hardware or software (as a “soft synth,” also called a “virtual instrument”). A MIDI sequence also can play samples: digital recordings
of musical notes played by real instruments.
Like a player piano, MIDI sequencing records your performance gestures rather than audio. Figure 2.7 shows the process.
MIDI/digital-audio software lets you record MIDI sequences and digital audio on hard disk. First record a few tracks of MIDI sequences,
then add audio tracks: lead vocal, sax solo, or whatever. All these elements stay synchronized.
PROS AND CONS OF EACH METHOD
Live stereo recording is simple, cheap, and fast. But it usually sounds
too muddy with rock music, and you must adjust balances by moving musicians. It can work well with classical music, and sometimes
with folk or acoustic jazz music.
Live mix recording is fairly simple and quick. However, loud instruments might sound distant in the recording because their sound
“leaks” into other instruments’ mics. You can’t correct mistakes in
11
12
Chapter 2 The Recording Chain
RECORDING
MIDI
signal
SOUND MODULE
Audio
signal
SEQUENCER
(MIDI recorder)
MIDI
signal
CONTROLLER
PLAYBACK
SEQUENCER
(MIDI recorder)
MIDI
signal
SOUND MODULE
Audio
signal
FIGURE 2.7
MIDI sequencing system.
the mix or performance unless you re-record. Also, the live sound of
the band can make it hard to hear the monitored sound clearly.
Multitrack recording has many advantages. You can punch-in: fix
a musical mistake on a track by recording a new, correct part over
the mistake. You can overdub: record one instrument at a time. This
reduces leakage and gives a tighter sound. Also, you can postpone
mixing decisions until after the performance. Then you can monitor
the mix in quiet surroundings. This method is more complex and
expensive than live-mix recording.
If you use a separate multitrack recorder and mixer, each component
can be used independently. For example, you might do a PA job with
just the mixer. Or, if you already have a mixer, all you need to buy
is a recorder. You’ll need to connect cables between the mixer and
recorder, and between the mixer and outboard effects units. It’s a
reliable, intuitive system for live recording or studio tracking.
A stand-alone DAW (recorder–mixer) is easy to use because it is a single
portable chassis that includes most of your studio equipment: recorder,
The Recording Chain Chapter 2
mixer, effects, and often a CD burner. It doesn’t require cables except
for the mics, instruments, and monitor speakers. High-end units let
you edit the music. They also have automated mixing: memory chips
in the mixer remember your mixdown settings, and reset the mixer
accordingly the next time you play back the recording.
A computer DAW is inexpensive, powerful, and flexible. It lets you do
sophisticated editing and automated mixing. Several plug-in (software) effects are included, and you can purchase and install other
plug-ins. Recording software can be updated at little cost. As for
drawbacks, computers can crash and can be difficult to set up and
optimize for audio work.
MIDI sequencing lets you record musical parts by entering notes
slowly or one at a time if you wish. After the sequencing is finished, you can edit notes to correct mistakes. You can even change
the instrument sounds or the tempo. A huge variety of sounds are
available in synthesizers, sound modules, and soft synths. However,
you’re limited to their sounds unless you use MIDI/digital-audio
software, which lets you add miked instruments to the mix.
The accompanying Web site contains samples of several types of
recordings. Play audio clips 1 and 2 at www.elsevierdirect.com/companions
/9780240811444. Clip 2 demonstrates:
■
■
■
■
■
Live stereo recording—orchestra
Live stereo recording—rock group
Live mix recording—jazz group
Multitrack recording—pop group
MIDI sequencing—synthesizer funk
RECORDING THE MIXES
No matter which recording method you use, eventually you’ll mix
each song and record the mix on a 2-track recorder, or record the mix
as a stereo WAVE file on your hard drive. You can convert the WAVE
file to an MP3 or WMA format for uploading to the Web (explained
in Chapter 20).
You might want to assemble an album of your recorded mixes.
To do this remove noises and count-offs between songs, put the
songs in the desired order, and put a few seconds of silence between
13
14
Chapter 2 The Recording Chain
songs. This is done with a computer and editing software (a DAW,
Figure 2.6). The last step is to copy the album to a blank CD. There’s
your final product, ready to duplicate or replicate.
No matter what type of recording you do, each stage contributes to
the sound quality of the finished recording. A bad-sounding CD can
be caused by any weak link: low-quality microphones, ineffective
mic placement or mixer settings, and so on. You can strive for quality
recordings by optimizing each stage. This book will help you reach
that goal.
CHAPTER 3
Sound, Signals, and Studio
Acoustics
When you make a recording, you deal with at least two kinds of
invisible energy: sound waves and electrical signals. For example, a
microphone converts sound into a signal. A signal is a varying voltage that carries information. In our case, it’s musical information.
This chapter covers some characteristics of sound and audio signals.
These facts will help you work with room acoustics, and will help
you know what’s going on inside your mixer as you adjust the controls. With this knowledge you can make better recordings.
SOUND WAVE CREATION
To produce sound, most musical instruments vibrate against air molecules, which pick up the vibration and pass it along as sound waves.
When these vibrations strike your ears, you hear sound.
To illustrate how sound waves are created, imagine a vibrating
speaker cone in a guitar amp. When the cone moves out, it pushes
the adjacent air molecules closer together. This forms a compression.
When the cone moves in, it pulls the molecules farther apart, forming a rarefaction. As shown in Figure 3.1, the compressions have a
higher pressure than normal atmospheric pressure; the rarefactions
have a lower pressure than normal.
These disturbances pass from one molecule to the next in a springlike motion—each molecule vibrates back and forth to pass the wave
along. The sound waves travel outward from the sound source at
1130 feet per second (344 meters per second), which is the speed of
sound in air at room temperature.
15
CHAPTER 3 Sound, Signals, and Studio Acoustics
SOUND
SOURCE
COMPRESSION
RAREFACTION
COMPRESSION
FIGURE 3.1
A sound wave.
PEAK
ABOVE NORMAL
AIR PRESSURE
16
TIME
NORMAL
BELOW NORMAL
TROUGH
FIGURE 3.2
Sound pressure vs. time of one cycle of a sound wave.
At some receiving point, such as an ear or a microphone, the air
pressure varies up and down as the disturbances pass by. Figure 3.2
is a graph showing how sound pressure varies with time, like a wave.
The high point of the graph is called a peak; the low point is called
a trough. The horizontal center line of the graph is normal atmospheric pressure.
Sound waves tend to spread out as they travel away from the source.
The compressions and rarefactions move out as expanding spheres.
Sound, Signals, and Studio Acoustics CHAPTER 3
ONE PERIOD
IN TIME
PEAK
AMPLITUDE
ONE CYCLE
ONE CYCLE
ONE CYCLE
FIGURE 3.3
Three cycles of a wave.
As the spherical waves expand, the sound pressure spreads over a
larger area, so the pressure becomes weaker with distance from the
source. That means, the farther you’re from a sound source, the
quieter its sound. Specifically, the sound pressure halves (drops
6 decibels; dB) each time the distance from the source doubles. This
phenomenon is called the inverse square law.
CHARACTERISTICS OF SOUND WAVES
Figure 3.3 shows three waves in succession. One complete vibration
from normal to high to low pressure and back to the starting point
is called one cycle. The time it takes to complete one cycle—from the
peak of one wave to the next—is called the period of the wave. One
cycle is one period long.
Amplitude
The height of the wave is its amplitude. Loud sounds have high
amplitudes (big pressure changes); quiet sounds have low amplitudes (small pressure changes). Play track 3 on the Web site www
.elsevierdirect.com/companions/9780240811444 to hear an example.
Frequency
The sound source (in this case, the guitar-amp loudspeaker) vibrates
back and forth many times a second. The number of cycles completed
in one second is called frequency. The faster the speaker vibrates, the
higher the frequency of the sound. Frequency is measured in hertz (Hz),
17
18
CHAPTER 3 Sound, Signals, and Studio Acoustics
which stands for cycles per second. One thousand hertz is called “one
kilohertz,” abbreviated kHz.
The higher the frequency, the higher the perceived pitch of the sound.
Low-frequency tones have a low pitch (like low E on a bass, which
is 41 Hz). High-frequency tones have a high pitch (like four octaves
above middle C, or 4186 Hz). Track 4 on the Web site www.elsevierdirect
.com/companions/9780240811444 illustrates this. Doubling the frequency
raises the pitch one octave.
Children can hear frequencies from 20 Hz to 20 kHz, and most adults
with good hearing can hear up to 15 kHz or higher. Each musical
instrument produces a range of frequencies, say, 41 Hz to 9 kHz for a
string bass, or 196 Hz to 15 kHz for a violin.
Wavelength
When a sound wave travels through the air, the physical distance
from one peak (compression) to the next is called a wavelength (refer
to Figure 3.1). Low-pitched sounds have long wavelengths (several
feet); high-pitched sounds have short wavelengths (a few inches or
less). Wavelength is the speed of sound divided by frequency. So the
wavelength of a 1000-Hz wave is 1.13 feet (0.344 meters); 100 Hz is
11.3 feet (3.44 meters), and 10 kHz is 1.35 inches (3.45 centimeters).
Phase and Phase Shift
The phase of any point on the wave is its degree of progression in the
cycle—the beginning, the peak, the trough, or anywhere between.
Phase is measured in degrees, with 360 degrees being one complete
cycle. The beginning of a wave is 0 degrees; the peak is 90 degrees
(one-quarter cycle), and the end is 360 degrees. Figure 3.4 shows the
phases of various points on the wave.
If there are two identical waves traveling together, but one is delayed
with respect to the other, there is a phase shift between the two
waves. The more delay, the more phase shift. Phase shift is measured in degrees. Figure 3.5 shows two waves separated by 90 degrees
(one-quarter of a cycle) of phase shift. The dashed wave lags the solid
wave by 90 degrees.
If you combine two identical sound waves, such as a sound and its
reflection off a wall, the peaks of the two waves add together at certain
Sound, Signals, and Studio Acoustics CHAPTER 3
AMPLITUDE
0
180 270
PHASE ANGLE
90
360
FIGURE 3.4
The phases of various points on a wave.
90
PHASE SHIFT
BETWEEN WAVES
AMPLITUDE
0
90 180 270 360
PHASE ANGLE OF SOLID WAVE
FIGURE 3.5
Two waves that are 90 degrees out-of-phase.
points in the room. This doubles the sound pressure or amplitude,
creating areas of louder sound at certain frequencies.
Phase Interference
When there is a 180-degree phase shift between two identical waves,
the peak of one wave coincides with the trough of another (Figure
3.6). If these two waves are combined, they cancel out. This phenomenon is called phase cancellation or interference.
Suppose you have a signal with a wide range of frequencies, such as the
singing voice. If you delay this signal and combine it with the original
undelayed signal, some frequencies will be 180 degrees out-of-phase
and will cancel. This makes a hollow, filtered tone quality.
19
20
CHAPTER 3 Sound, Signals, and Studio Acoustics
AMPLITUDE
TIME
AMPLITUDE
DELAY CAUSING
A 180-DEGREE
PHASE SHIFT
TIME
AMPLITUDE
SUM OF BOTH WAVES ABOVE
TIME
FIGURE 3.6
Phase interference: adding two waves that are out-of-phase cancels the sound at that frequency.
Here’s an example of how this can happen. Suppose you’re recording a singer/guitarist with one mic near the singer and another mic
near the guitar. Both mics pick up the singer. The singer’s mic is close
to the mouth, and you hear it with no delay in the signal. The guitar
mic is farther from the mouth, so its voice signal is delayed. When
you mix the two mics, you often hear a colored tone quality caused
by phase cancellations between the two mics.
Suppose you’re recording a stage play with a mic on a short stand
on the floor. The mic picks up the direct sound from the actors, but
it also picks up delayed reflections off the floor. Direct and delayed
sounds combine at the mic, causing phase cancellations. You hear it
as a hollow, filtered sound that changes when the actor walks while
talking.
Sound, Signals, and Studio Acoustics CHAPTER 3
FUNDAMENTAL
+
SECOND
HARMONIC
+
THIRD
HARMONIC
=
RESULTING COMPLEX
WAVEFORM (SUM OF THE
THREE WAVES ABOVE)
FIGURE 3.7
Adding fundamental and harmonic waveforms to form a complex waveform.
Harmonics
The type of wave shown in Figure 3.2 is called a sine wave. It is a pure
tone of a single frequency, like a signal from a tone generator. In contrast,
most musical tones have a complex waveform, which has more than
one frequency component. All sounds are combinations of sine waves
of different frequencies and amplitudes. Figure 3.7 shows sine waves of
three frequencies combined to form a complex wave.
The lowest frequency in a complex wave is called the fundamental
frequency. It determines the pitch of the sound. Higher frequencies
in the complex wave are called overtones or upper partials. If the
overtones are multiples of the fundamental frequency, they are called
harmonics. For example, if the fundamental frequency is 200 Hz, the
second harmonic is 400 Hz, and the third harmonic is 600 Hz.
Harmonics and their amplitudes partly determine the tone quality
or timbre of a sound. They help you identify the sound as a drum,
piano, organ, voice, etc. Play audio clip 5 at www.elsevierdirect.com/
21
22
CHAPTER 3 Sound, Signals, and Studio Acoustics
companions/9780240811444. In general, instruments with few or
weak harmonics—such as a flute—tend to sound pure and smooth.
Instruments with many or strong harmonics—such as a trumpet or
distorted guitar—tend to sound bright and edgy.
Explained in Chapter 10, equalization can change the tonal balance
of a recorded instrument by boosting or cutting its harmonics and
fundamental frequencies. Boosting the fundamentals tends to make
the sound warm; cutting the fundamentals makes the sound thin;
boosting the harmonics makes the sound bright, defined, or trebly;
and cutting the harmonics makes the sound dark or muffled.
Playing an instrument louder usually increases its harmonics. So a
piano played loudly sounds brighter than the same piano played
softly.
Noise (such as tape hiss) contains a wide band of frequencies and
has an irregular, nonrepeating waveform. Some musical sounds with
a hissy or noise-like character are a cymbal, snare drum, or the “s”
sounds a singer makes (sibilance).
Envelope
Another characteristic that identifies a sound is its envelope. When a
note sounds, it doesn’t last forever—it rises in volume, lasts a short
time, then falls back to silence. This rise and fall in volume of one
note is called the note’s envelope. The envelope connects the peaks
of successive waves that make up a note. Each musical instrument
has a different envelope.
Most envelopes have four sections: attack, decay, sustain, and release
(see Figure 3.8). During the attack, a note rises from silence to its
maximum volume. Then it decays from maximum to some midrange level. This middle level is the sustain portion. During release,
the note falls from its sustain level back to silence.
Percussive sounds, such as drum hits, are so short that they have
only a rapid attack and decay. Other sounds, such as organ or violin notes, last longer. They have slower attacks and longer sustains.
Guitar plucks and cymbal crashes have quick attacks and slow
releases. They hit hard then fade out slowly. Play audio clip 6 at www
.elsevierdirect.com/companions/9780240811444.
Sound, Signals, and Studio Acoustics CHAPTER 3
ATTACK
DECAY
RELEASE
PEAK
AMPLITUDE
SUSTAIN
TIME
FIGURE 3.8
The four sections of the envelope of a note.
You can shorten a guitar string’s decay or ringing by damping the
string with the side of your hand. You can press a blanket against a
kick drum head to damp the decay and get a tighter sound.
Harmonics usually change during a note’s envelope. For example
if an instrument has a percussive attack—like a guitar pluck or tom
hit—the harmonics are strongest at the attack then become weaker
during the decay.
BEHAVIOR OF SOUND IN ROOMS
Because most music is recorded in rooms, you need to understand
how room surfaces affect sound.
Echoes
Musical instruments make sound waves that travel outward in all
directions. Some of the sound travels directly to your ears (or to a
microphone) and is called direct sound. The rest strikes the walls,
ceiling, floor, and furnishings of the recording room. At those surfaces, some of the sound is absorbed, some is transmitted through
the surface, and the rest is reflected back into the room.
Because sound waves take time to travel (moving at about 1 foot per
millisecond), the reflected sound reaches you after the direct sound.
The reflection repeats the original sound after a short delay. If the
23
24
CHAPTER 3 Sound, Signals, and Studio Acoustics
ROOM
SURFACE
REFLECTED SOUND
ECHO
INSTRUMENT
DIRECT
SOUND
MICROPHONE
(A)
DIRECT-SOUND
IMPULSE
ECHO
FLUTTER ECHOES
LEVEL
0
(B)
50 100 150 200 250
TIME IN MILLISECONDS
FIGURE 3.9
Echoes: (A) echo formation; (B) intensity vs. time of direct sound and its echoes.
sound is delayed about 50 milliseconds or more, we call it an echo
(Figure 3.9). In some concert halls we hear single echoes; in small
rooms we often hear a short, rapid succession of echoes called flutter
echoes. You can detect them by clapping your hands next to a wall.
Flutter echoes happen when sound bounces back and forth between
two parallel walls.
Reverberation
Sound reflects many times from all the surfaces in the room. These
reflections sustain the sound of each note the musician plays. This
persistence of sound in a room after the original sound has stopped
is called reverberation (reverb). For example, reverberation is the
sound you hear just after you shout in an empty gymnasium. The
sound of your shout stays in the room and gradually dies away
(decays). Play audio clip 10 at www.elsevierdirect.com/companions
/9780240811444 to hear echoes and reverberation.
Sound, Signals, and Studio Acoustics CHAPTER 3
CEILING
SOUND
REFLECTIONS
WALL
DIRECT
SOUND
MICROPHONE
FLOOR
(A)
DIRECT-SOUND
IMPULSE
EARLY
REFLECTIONS
REVERBERATION
LEVEL
(B)
0
100 200 300 400 500
TIME IN MILLISECONDS
FIGURE 3.10
Reverberation: (A) multiple sound reflections create reverberation; (B) intensity vs. time of direct sound,
early reflections, and reverberation.
Reverb is hundreds of echoes that gradually get quieter. The echoes
follow each other so rapidly that they merge into a single continuous
sound. Eventually, the room surfaces completely absorb the echoes.
The timing of the echoes is random, and the echoes increase in number as they decay. Figure 3.10 shows how reverberation develops in a
recording room.
Reverberation is a continuous fade-out of sound (“HELLO-O-O-o-o”),
while an echo is a discrete repetition of a sound (“HELLO hello hello
hello”).
Reverberation time (RT60) is the time it takes for reverb to decay
60 dB. Too long a reverb time makes a recording sound distant,
muddy, and washed-out. That’s why pop-music recordings are usually
made in a fairly “dead” or nonreverberant studio, which has an RT60
of about 0.4 second or less. In contrast, classical music is recorded in
“live,” reverberant concert halls (RT60 about 1 to 3 seconds) because
we want to hear reverb with classical music—it’s part of the sound.
25
26
CHAPTER 3 Sound, Signals, and Studio Acoustics
Reverberation comes to you from every direction because it is a pattern of many sound reflections off the walls, ceiling, and floor.
Because we can tell where sounds come from, we can distinguish
between the direct sound of an instrument coming to us from a single location and the reverberation coming to us from everywhere else.
So we can ignore the reverb and concentrate on the sound source. In
fact, we normally aren’t aware of reverberation.
But suppose you put a mic next to your ears, record an instrument
in that room, and play back the recording. You’ll hear a lot more
reverb than what you heard live. What’s going on? The reverb you
recorded and played back isn’t all around you. Instead, it’s all up
front between the speakers. So it’s more audible; you can’t discriminate against the reverb spatially. To reduce the amount of reverb in
your recordings, you need to place mics close to instruments, and
maybe add some sound-absorbing materials to the room.
Diffusion
When sound waves strike and reflect off a very bumpy or convex surface, they spread out or diffuse. This diffusion can be used to weaken
sound reflections. Sound waves also spread out when they travel
through a small opening.
HOW TO TAME ECHOES AND REVERB
Echoes and reverb can make your recordings sound mushy and distant. There are two ways to prevent these problems: with recording
techniques and with acoustic treatment.
Controlling Room Problems with Recording Techniques
Sometimes you can make clean recordings in an ordinary room, such
as a club, living room, or basement, if you follow these suggestions:
■
■
Mike close. Place each mic 1 to 6 inches from each instrument
or voice. Then the mics will hear more of the instruments and
less of the room reflections. You might want to use mini mics,
which attach directly to instruments.
Use directional mics—cardioid, supercardioid, or hypercardioid—
which reject room acoustics.
Sound, Signals, and Studio Acoustics CHAPTER 3
■
Record bass guitar and synth directly with a guitar cord or a
direct box. Since you omit the microphone, you pick up no
room acoustics. To get a good sound when recording electric
guitar direct, record off the effects boxes or use a guitar-amp
simulator.
Also see the tips on reducing leakage in Chapter 7 under the heading, Leakage (Bleed or Spill).
Controlling Room Problems with Acoustic Treatments
When should you apply acoustic treatment to a room or build a
studio?
■
■
■
■
■
■
■
You clap your hands next to a wall and you hear flutter echoes
(a fluttering sound). These are caused by sounds bouncing back
and forth between hard parallel walls.
Your studio is a very live environment, such as a garage or concreteblock basement, so you hear too much room reverberation.
Your studio is very small.
You hear outside noises in your recordings.
Bass-guitar amps and monitor speakers sound boomy.
You want the freedom to mike several feet away without picking
up noise or excess room reverb.
You hear a lot of leakage in the mic signals. Examples of leakage
are drums picked up by the guitar mic or electric guitar picked
up by the cymbal mics.
If those conditions apply, upgrade the acoustics of your studio. Here
are some suggestions.
Reverb and echoes are caused by sound reflections off room surfaces.
So any surface that is highly sound-absorbent helps to reduce those
problems.
To absorb high frequencies, use porous materials such as convoluted
(bumpy) foam mattresses. They can be highly flammable, so cover
them with flame-retardant treatment (see natfire.com or flamestop
.com). Nail or glue them to the walls, or mount them on frames.
Thick foam works better than thin. Four-inch foam on the wall
absorbs frequencies from about 200 to 800 Hz and up, depending on
27
28
CHAPTER 3 Sound, Signals, and Studio Acoustics
MONITORS
WEDGE-SHAPED
FOAM, OR PRESSED
FIBERGLASS
INSULATION
BOARDS, REDUCE
REVERB AND
FLUTTER ECHOES
MIXER
TUBULAR BASS TRAPS
REDUCE STANDING WAVES,
ABSORB LOWS
PLACE THEM
1/4 WAVELENGTH
FROM WALL
UNEQUAL ROOM DIMENSIONS
REDUCE STANDING WAVES
PRESSED FIBERGLASS
INSULATION BOARDS
ACROSS CORNERS
REDUCE STANDING WAVES
DIFFUSORS REDUCE
FLUTTER ECHOES
INSULATION-FILLED
CORNER BEHIND AN
OPEN LATTICE IS
ANOTHER TYPE OF
BASS TRAP
FIGURE 3.11
Acoustic treatments.
the angle at which sound strikes the foam. Leave some space between
the foam panels (Figure 3.11); this helps to diffuse or spread out the
sound in the room. Don’t overdo the foam padding. A stuffed, dead
room is uncomfortable to play in. Keep some reflections because they
add “air” and liveliness to the sound.
Other high-frequency absorbers are sleeping bags, moving blankets,
carpeting, curtains, and dense fiberglass insulation covered with
muslin or burlap. If possible, space these materials several inches
from the wall. The spacing helps them absorb mid-bass frequencies.
A wide-range absorber is 4-inch pressed fiberglass board (OwensCorning Type 703, 3 lb/ft3) covered in muslin or burlap.
Start with just a little absorption behind or above the musician you’re
recording. Add more absorbers a few at a time, until your recordings
sound as dead as you wish—typically covering 50–60% of the surfaces. Place absorption where you see mirror images of the monitors
from the mix position. Put an absorber on the wall behind the monitors, and also hang an absorbent panel centered halfway between the
mix position and monitors, suspended on hooks and wire.
Sound, Signals, and Studio Acoustics CHAPTER 3
Another absorber is a stand-mounted acoustical panel placed near a
microphone. Examples are the ModTrap and the sE Relexion Filter.
To absorb low frequencies, you can make bass traps. Here are four
types:
1. Resonant tube trap: Take a 35- to 55-gallon rubber trashcan, stuff it with fiberglass insulation (wear a dust mask and
gloves), and cover the open end with muslin or canvas. This
tube trap absorbs frequencies near 1130/2 H, where H is the
height of the trashcan in feet. For example, a 3-foot-high trashcan absorbs 188 Hz. Placement is not critical.
2. Frictional tube trap: Make a 2-foot-diameter canvas bag, 8 feet
tall, and fill it with fiberglass insulation. Hang one bag a few
feet out from each room corner (Figure 3.11). The distance
should be 1130/4F, where F is the frequency in Hz that you
want to absorb. For example, to absorb 80 Hz, hang the bag
3.5 feet out from a room corner. (Thanks to David Moulton for
these ideas).
3. Lattice: Build a 3-foot-wide flat lattice frame of wood slats, the
same height as the ceiling (Figure 3.11). Cover the frame with
muslin or burlap. Put the frame diagonally across a room corner and fill the corner with R-30 insulation. Leave the foil on,
foil side out toward the room. Put one assembly in each corner. (Thanks to Chips Davis for this idea).
4. Insulation panel: Get 8 pieces of rigid fiberglass insulation,
type 705, 2 ⫻ 4 feet and 4 inches thick, from an insulation
supplier. Cover each piece with muslin or burlap to contain the
fibers. Place a piece across each room corner with the 2-foot
edge touching the floor. Stack two panels to make them 8 feet
high. (Thanks to Ethan Winer for this idea).
There are other ways to absorb bass. Wood paneling works well. It
also helps to open closet doors and place couches and books a few
inches from the walls. In a basement studio, nail acoustic tile to the
ceiling joists with fiberglass insulation in the air space between tiles
and ceiling.
You may not need any bass traps if you don’t put any bass into the
room. For example, don’t turn up the bass-guitar amp—just record the
bass direct and have the musicians wear headphones to hear the bass.
29
30
CHAPTER 3 Sound, Signals, and Studio Acoustics
Controlling Standing Waves
Let’s look at another acoustic problem: standing waves. If you play
an amplified bass guitar through a speaker in a room, and do a bass
run up the scale, you may hear some notes that boom out in the
room. The room is resonating at those frequencies. These resonance
frequencies, which are strongest below 300 Hz, occur in patterns
called “standing waves.” They can give a tubby or boomy coloration
to musical instruments and monitor speakers.
Room resonances are worst in a cubical room. They are less of a
problem if the room’s length, width, and height are not multiples
of each other. Table 3.1 shows several room dimensions in feet that
reduce boomy-sounding standing waves.
For example, if the room width is 9.1 feet, and the ceiling is 8 feet
high, the length should be 11.1 feet for best reduction of boominess.
Try to record in a large room because the room resonance frequencies are likely to be below the musical range. Use bass traps to absorb
Table 3.1
Height
Room Dimensions in Feet to Reduce Standing Waves
Width
Length
8
9.1
11.1
8
9.4
11.8
8
10.1
11.3
8
10.2
12.3
8
11.6
16.8
8
11.8
13.6
8
12.8
18.6
8
13.0
21.0
10
11.4
13.9
10
11.7
14.7
10
12.6
14.1
10
12.8
15.4
10
14.5
21.0
10
14.7
17.0
10
16.0
23.3
10
16.2
26.2
Sound, Signals, and Studio Acoustics CHAPTER 3
room resonances. Contrary to popular opinion, nonparallel walls
don’t prevent standing waves.
MAKING A QUIETER STUDIO
Ideally the noise level in a studio should be 23–28 dB A-weighted as
measured on an SPL meter. The following tips will keep noises out
of your recordings:
■
■
■
■
■
■
■
■
■
■
■
Consider having the studio in a basement, because the earth
blocks noises from the outside. The furnace or air-conditioning
might be a problem, however.
Turn off appliances, air conditioning, and telephones while
recording.
Pause for ambulances and airplanes to pass.
Close windows. Consider covering them with thick plywood.
Close doors and seal with towels.
Remove small objects that can rattle or buzz.
Weather-strip doors all around, including underneath.
Replace hollow doors with solid doors.
Block openings in the room with thick plywood and caulking.
Put several layers of plywood and carpet on the floor above the
studio, and put insulation in the air space between the studio
ceiling and the floor above.
Place microphones close to instruments and use directional
microphones. This won’t reduce noise in the studio, but it will
reduce noise picked up by the microphones.
When building a new studio, you might want to make the walls of
plastered concrete block (because massive walls reduce sound transmission) or make the walls of gypsum board and staggered studs. Nail
gypsum board to 2 ⫻ 4 staggered studs on 2 ⫻ 6 footers as shown in
Figure 3.12. Staggering the studs prevents sound transmission through
the studs. Fill the air space between walls with insulation.
The ideal home-recording room for pop music is a large, well-sealed
room with optimum dimensions. This room is in a quiet neighborhood. It should have some soft surfaces (carpet, acoustic-tile ceiling,
drapes, couches), and some hard vibrating surfaces (wood paneling
or gypsum board walls on studs).
Your home studio may not need acoustic treatment. Do some trial
recordings to find out. But if your room could stand some improvement, the suggestions above should point you in the right direction.
31
32
CHAPTER 3 Sound, Signals, and Studio Acoustics
GYPSUM BOARD
2X4 STUDS ON 2X6 FOOTER
6"
8"
8"
FIBERGLASS
INSULATION
FIGURE 3.12
Staggered-stud construction reduces noise transmission.
For better results and a more professional appearance, consider buying some acoustic treatments from the following companies. Their
Web sites are atsacoustics.com, realtraps.com, acousticsciences.com,
acousticalsolutions.com, primacoustic.com, auralex.com, acousticsfirst
.com, wallmate.net, pinta-acoustic.com, and rpginc.com. Atsacoustics
.com has a free online room acoustics analysis.
SIGNAL CHARACTERISTICS OF AUDIO DEVICES
When a microphone converts sound to electricity, this electricity is
called the signal. It has the same frequencies and the same amplitude changes as the incoming sound wave.
When this signal passes through an audio device, the device may
alter the signal. It might change the level of some frequencies or add
unwanted sounds that aren’t in the original signal. Let’s look at some
of these effects.
Frequency Response
Suppose you have an audio device—a mic, mixer, effects unit,
recorder, or speaker. You send a musical signal through the device.
Usually the music contains some high and some low frequencies.
The device might respond differently to different frequencies. It might
amplify the low notes and weaken the high notes. You can graph
how the device responds to different frequencies by plotting its output level versus frequency. This graph is called a frequency response
(Figure 3.13). The level in the graph is measured in dB, while frequency is measured in Hz. Generally, 1 dB is the smallest change in
loudness that we can hear.
RELATIVE RESPONSE IN dB
Sound, Signals, and Studio Acoustics CHAPTER 3
+15
+10
+5
0
–5
–10
–15
20
50
100
500
1K
5K
10K
20K
5K
10K
20K
FREQUENCY in Hz
RELATIVE RESPONSE IN dB
FIGURE 3.13
An example of a frequency response.
+15
+10
+5
0
–5
–10
–15
20
50
100
500
1K
FREQUENCY in Hz
FIGURE 3.14
Flat frequency response.
Suppose the level is the same at all frequencies. Then the graph
forms a horizontal straight line and is called a “flat frequency
response” (Figure 3.14). All frequencies are reproduced at an equal
level. In other words, the device passes all the frequencies without
changing their relative levels. You get out the same amount of bass
and treble that went in. A flat response doesn’t affect the tonal balance of the incoming sound.
Many audio devices don’t have a flat response across the audio band
from 20 to 20,000 Hz. They have a limited range of frequencies that
can be reproduced at an equal level (within a tolerance, such as
⫾3 dB). In Figure 3.14, the frequency response shown by the solid
line is 50 to 12,000 Hz ⫾ 3 dB. That means the audio device passes
all frequencies from 50 to 12,000 Hz at a nearly equal level—within
3 dB. It reproduces low and high sounds equally well. The response is
down 3 dB at 50 and 12,000 Hz, and is up 3 dB at 5000 Hz.
Usually, the more extended or “wide” the frequency range is, the
more natural and real the recording sounds. A wide, flat response
33
34
CHAPTER 3 Sound, Signals, and Studio Acoustics
gives accurate reproduction. A frequency response of 200 to 8000 Hz
(⫾3 dB) is narrow (poor fidelity); 80 to 12,000 Hz is wider (better
fidelity), and 20 to 20,000 Hz is widest (best fidelity). Play audio clip
7 at www.elsevierdirect.com/companions/9780240811444.
Also, the flatter the frequency response, the greater the fidelity or
accuracy. A response deviation of ⫾3 dB is good, ⫾2 dB is better,
and ⫾1 dB is excellent. There are exceptions to this statement, which
we’ll look at in Chapter 10 in the section on equalizers.
When you turn a bass or treble knob on your guitar amp, mixer EQ,
or stereo, you’re changing the frequency response. If you turn up the
bass, the low frequencies rise in level. If you turn up the treble, the
high frequencies are emphasized. The ear interprets these effects as
changes in tone quality—warmer, brighter, thinner, duller, and so on.
Figure 3.13 shows a non-flat frequency response. Toward the right
side of this line, the response at high frequencies “rolls off” or
declines. This shows that the upper harmonics are weak. The result
is a dull sound. Toward the left side, the response at low frequencies
rolls off. This means the fundamentals are weakened and the result
is a thin sound.
The frequency response of an audio device might be made non-flat
on purpose. For example, you might cut low frequencies with an
equalizer to reduce breath pops from a microphone. Also, a microphone may sound best with a non-flat response, such as boosted
high frequencies that add presence and sizzle.
Noise
Noise is another characteristic of audio signals. Every audio component produces a little noise—a rushing sound like wind in trees.
Noise in a recording is undesirable unless it’s part of the music.
You can make noise less audible by keeping the signal level in a
device relatively high. If the level is low, you have to turn up the
listening volume to hear the signal well. Turning up the volume of
the signal also turns up the volume of the noise, so you hear noise
along with the signal. But if the signal level is high, you don’t have
to turn up the listening level as much. Then the noise remains in the
background.
Sound, Signals, and Studio Acoustics CHAPTER 3
Distortion
If you turn up the signal level too high, the signal distorts and you
hear a gritty, grainy sound or clicks. This type of distortion is sometimes called “clipping” because the peaks of the signal are clipped
off so they are flattened. To hear distortion, simply record a signal
at a very high recording level (with the meters going well into the
red area) and play it back. Play audio clip 9 at www.elsevierdirect.com/
companions/9780240811444. Digital recorders also produce “quantization” distortion at very low signal levels.
Optimum Signal Level
You want the signal level high enough to cover up the noise, but low
enough to avoid distortion. Every audio component works best at a
certain optimum signal level, and this is usually indicated on a level
meter built into the device. More on this in Appendix A.
Figure 3.15 shows the range of signal levels in an audio device. At the
bottom is the noise floor of the device—the level of noise it produces
with no signal. At the top is the distortion level—the point at which
the signal distorts and sounds grungy. In between is a range in which
the signal sounds clean. The idea is to maintain the signal around normal operation level on the average. Generally you want the signal level
to be as high as possible without clipping or distorting.
Distortion (clipping) level
Signal
level
Headroom
Normal
operation
level
Level
Signal-to-noise
ratio
Noise floor
Time
FIGURE 3.15
The range of signal levels in an audio device.
35
36
CHAPTER 3 Sound, Signals, and Studio Acoustics
With digital recorders (such as computer recording software),
“0 dBFS” (0 decibels full scale) on the meter is the maximum undistorted signal. In that case, an average level of ⫺20 dBFS and a peak
level of ⫺6 dBFS is the normal operation level. In a digital audio file
that is used to master a CD, an average level of ⫺14 dBFS is a good
goal to aim for. There’s more on digital signal levels in Chapter 13
on Computer Recording.
Signal-to-Noise Ratio
The level difference in decibels between the signal level and the noise
floor is called the signal-to-noise ratio or S/N (see Figure 3.15). The
higher the S/N, the cleaner the sound. An S/N of 60 dB is fair, 70 dB
is good, and 80 dB or greater is excellent. Play audio clip 8 at www
.elsevierdirect.com/companions/9780240811444.
To illustrate S/N, imagine a person yelling a message over the sound
of a train. The message yelled is the signal; the noise is the train. The
louder the message, or the quieter the train, the greater the S/N. And
the greater the S/N, the clearer the message.
Headroom
The level difference in dB between the normal signal level and the
distortion level is called “headroom” (see Figure 3.15). The greater
the headroom, the greater the signal level the device can pass without running into distortion. If an audio device has a lot of headroom, it can pass high-level peaks without clipping them.
You want to set your mixer controls so that the signal has some
headroom, is well above the noise floor, and is below distortion. Tips
on doing that are in Chapter 9, Digital Recording, under the heading Digital Recording Level. Also see Chapter 12, Mixer Operation,
under the headings Set Recording Levels and Set Levels.
With these settings, the signal levels in the mixer should be just
about right, with no audible noise or distortion. And the mixer
should have enough headroom so that loud peaks won’t distort.
CHAPTER 4
Equipping Your Studio
You want to set up a recording system that’s affordable, easy to use,
and sonically excellent. With today’s wide array of user-friendly sound
tools, you can do just that. This chapter is a guide to equipment for a
recording studio: what it does, what it costs, and how to set it up.
In this chapter we’ll examine:
■
■
■
Equipment
Cables and connectors
Preventing hum and radio frequency interference
What is the bare-bones equipment you need to crank out a decent
demo CD? How much does it cost? Thanks to the new breed of
affordable equipment, you can put together a complete home recording studio for as little as $1200. That includes powered speakers, two
mics and mic stands, recording software, a sound card, headphones,
and cables.
EQUIPMENT
Let’s examine each piece of equipment in a recording studio. You’ll
need some sort of recording device, headphones, cables, mics, direct
box, monitor speakers, audio interface, and effects.
Recording Device
Five types of recording devices to choose from are a 2-track recorder,
a recorder–mixer, a separate multitrack recorder and mixer, a computer, and a keyboard workstation. We’ll look at each one.
37
38
CHAPTER 4 Equipping Your Studio
PORTABLE 2-TRACK RECORDER
This device is great for on-location stereo recording of classical music:
an orchestra, symphonic band, string quartet, pipe organ, or soloist.
Sometimes it can be used to record folk music or jazz, or to capture a
practice session or an audience perspective of a rock group. Two types
of 2-track recorder are a Flash memory recorder (Figure 4.1), or laptop
computer with recording software. Prices range from $200 to $1800.
Details on these units are in Chapter 9. Some Flash recorders are the
Sony PCM-D1 and D50 (sony.com/proaudio), Marantz PMD620
(d-mpro.com), Edirol R-1, R-09 and R-09HR (edirol.com), M-Audio
MicroTrack II (m-audio.com), Zoom H4 and H2 (samsontech
.com), Olympus LS-10 (olympus.com), Korg MR-1 and MR-1000
(korg.com), Yamaha PockeTrak 2G (yamaha.com), and TASCAM
DR-1 and HD-P2 (tascam.com).
FIGURE 4.1
Edirol R-09W, an example of a portable 2-track recorder.
Equipping Your Studio CHAPTER 4
MINI STUDIO
Also called a portable studio, personal studio, or pocket studio, this
piece of gear combines a 4- or 8-track recorder with a mixer—all in
one portable chassis (Figure 4.2). A mini studio records in MP3 format to a Flash memory card. It lets you record several instruments and
vocals, then mix them to stereo; that is, you “bounce” the tracks to a
stereo file in the memory card. You can copy the stereo mix file to your
computer via USB. (USB is a type of cable and ports for high-speed
data transfer). The sound quality of Pocket Studios is good enough to
make demos, or to use as a musical scratchpad for ideas, but it is not
quite good enough to release commercial albums. Costing about $230
to $599, a mini studio can be a good choice for beginning recordists.
Some features include:
■
■
An internal MIDI sound module (synthesizer) that plays back
MIDI sequences. Includes MIDI files or rhythm patterns to jam
with.
Built-in effects.
FIGURE 4.2
Boss BR-600, an example of a mini studio.
39
40
CHAPTER 4 Equipping Your Studio
■
■
■
■
■
■
Built-in mic (in some models).
Autopunch. Automatically “punch” into and out of record
mode to correct mistakes.
Battery or AC-adapter powered.
Virtual tracks let you record multiple takes of a single performance, then select your favorite take during mixdown.
Guitar-amp modeling simulates various guitar amps; mic modeling simulates mic models.
2 mic inputs; records up to 2 tracks at a time. Plays back 4 or 8
tracks at once.
DIGITAL MULTITRACKER (RECORDER–MIXER)
Other names for this device are stand-alone Digital Audio Workstation
(DAW), portable digital studio, or personal digital studio.
Like the mini studio, the digital multitracker combines a multitrack
recorder with a mixer in a single chassis (Figure 4.3). It’s convenient
and portable. Offering better sound than a mini studio, the digital
multitracker provides CD sound quality because it can record WAVE
files. It records 8 to 32 tracks onto a built-in hard drive or Flash
memory card. The mixer includes faders (volume controls) for mixing, EQ or tone controls, and aux sends for effects (such as reverb).
An LCD screen displays recording levels, waveforms for editing, and
other functions.
Price ranges from $700 to $3600. Some manufacturers of mini studios or digital multitrackers are TASCAM, Boss, Fostex, Korg, Roland,
Yamaha, and Zoom (Samson).
Chapter 9, under the heading Recorder–Mixer (Personal Digital
Studio), lists some features to look for in a recorder–mixer.
SEPARATE MULTITRACK RECORDER AND MIXER
Ideal for on-location recording, a multitrack hard-disk recorder (HD
Recorder) records up to 24 tracks reliably on a built-in hard drive
(Figure 4.4), just like the hard drive in your computer. The recorder
doesn’t have mic preamps, so it must be used with a mixer or mic
preamps. Multiple recorders can be linked to get more tracks. Some
examples are the Alesis ADAT HD24XR, TASCAM MX-2424 and
X-48, Otari DR-100, iZ Technology RADAR, Fostex D-2424LV, and
Mackie HDR 24/96. Prices range from $1500 to $20,000.
Equipping Your Studio CHAPTER 4
FIGURE 4.3
Korg D3200, an example of a digital multitracker (recorder–mixer).
FIGURE 4.4
Alesis HD24XR, an example of a multitrack HD recorder.
Current units record with 24-bit resolution and up to 96 kHz sampling rate. Some record 24 tracks of 24-bit programs at sample rates
of 44.1 and 48 kHz. They record 12 tracks of 24-bit recording at 88.2
and 96 kHz. (These terms are explained in Chapter 9). Some HD
recorders allow track editing, either on a built-in LCD screen or on
41
42
CHAPTER 4 Equipping Your Studio
a plug-in computer monitor. A few models have built-in removable
hard drives for backing up projects.
A mixer (Figure 4.5) is an electronic device with several mic preamplifiers, volume and tone controls for each mic, monitor volume controls,
and outputs for effects devices. To use it, plug mics and electric instruments into the mixer, which amplifies their signals. Connect the output
signal of each mic preamp to a multitrack recorder. While recording,
use the mixer to send those signals to the desired recorder tracks, and
to set recording levels. During mixdown, the mixer combines (mixes)
the tracks to stereo. It also lets you adjust each track’s volume, tone
quality, effects, and stereo position. A large, complex mixer is called a
FIGURE 4.5
Mackie 1604-VLZ3, an example of a mixer.
Equipping Your Studio CHAPTER 4
mixing console (board or desk) and costs upwards of $3000. Chapter
11 explains mixing consoles in detail.
Price is about $180 to $1500. Mixers are made by such companies
as TASCAM, Alesis, Yamaha, Mackie, Allen & Heath, Soundcraft,
M-Audio, Tapco, Peavey, Samson, and many others.
COMPUTER DIGITAL AUDIO WORKSTATION (DAW)
This low-cost recording setup has three parts: a personal computer, an
audio interface, and recording software (Figure 4.6). Some software
records MIDI data (explained later) as well as audio. The audio interface converts audio from your mics, mic preamps, or mixer into a signal that is recorded on the computer’s hard drive as a magnetic pattern.
Dozens of tracks can be recorded with professional quality. You mix
the tracks with a mouse by adjusting the “virtual” controls that appear
on your computer screen. Then you record the mix on your computer’s
hard drive. Details are in Chapter 13, Computer Recording.
MONITOR
HARD
DRIVE
PCI SLOT, USB PORT,
OR FIREWIRE PORT
COMPUTER
KEYBOARD
ALESIS
FIREPORT
AUDIO INTERFACE
SOUND CARD
AUDIO INTERFACES (CHOOSE ONE)
FIGURE 4.6
Computer with recording software and a choice of audio interface.
MOUSE
CONTROLLER OR MIXER
WITH I/O
43
44
CHAPTER 4 Equipping Your Studio
A computer studio costs about the same as a mini studio and is more
powerful. It’s a bargain. But since software requires computer skills,
it’s a little harder to learn and use than a hardware multitracker.
Software recordings are at least CD quality—better than the MP3
recordings you get with a mini studio.
Recording software costs anywhere from $0 (freeware) to $1800,
with $150 to $500 being typical prices. Software examples include
Cakewalk Home Studio, Sonar Studio and Sonar Producer;
Digidesign Pro Tools LE, M-Powered and HD; Apple Logic Studio
and Logic Pro, Steinberg Cubase and Nuendo, Sony Vegas and
Sound Forge, BIAS Deck, Pro Tracks Plus, Magix Samplitude and
Sequoia, Mackie Tracktion, Reaper; N Track Studio, RML Labs SAW
Studio, Adobe Audition, and MOTU Digital Performer (Mac only).
Free multitrack recording programs are available for download.
Although they lack extensive features, they offer a chance to practice
your skills at no cost. Examples are Audacity (http://audacity.sourceforge
.net), Kristal (from Kreative.org), and Garage Band (supplied free with
any new MacIntosh computer or $79 with a Mac software upgrade).
Related to recording software is music creation software, which comes
with samples of musical instruments. You set up loops (repeating
drum riffs and musical patterns), play along with them using a MIDI
keyboard, and record that MIDI sequence. Examples include Ableton
Live, Propellerhead Reason, Sony ACID Pro, and Spectrasonics
Stylus RMX.
FIGURE 4.7
M-Audio Fast Track Ultra, an example of a USB 2.0 audio interface. It offers low-latency direct monitoring
with reverb, and can create two headphone mixes.
Equipping Your Studio CHAPTER 4
A sound card (PCI audio interface) costs $100 to $1700, a USB
audio interface (Figure 4.7) costs $100 to $500, and a FireWire audio
interface costs $200 to $1950.
Some makers of sound cards include Echo, Event, Frontier Design,
M-Audio, RME, Lynx, and E-MU. USB and FireWire Audio interface
manufacturers include Digidesign, Metric Halo, MOTU, PreSonus,
Apogee, Focusrite, M-Audio, TC Electronic, Mackie, E-MU, Alesis,
Edirol, TASCAM, Lexicon, and many others.
Another type of audio interface is a mixer that connects to a computer by a FireWire or USB port. It sends the output signal of each
mic preamp to a separate track in your recording software, and
returns the stereo mix back to the mixer for monitoring. This can
be a good system for recording a band because it offers easy setup
of monitor mixes with effects. Some examples are the Mackie Onyx
1640 with the optional Onyx FireWire Card (16 recording channels),
Alesis Multimix USB or FireWire series (up to 8 recording channels),
Edirol M-16DX (16 recording channels, 4 mic inputs), and Behringer
Xenyx 2442FX (4 recording channels). Generally these mixers are not
controllers for recording software—you still need to use a mouse—
but the Edirol M-16DX is also a control surface.
A few microphones have USB built in, so you just plug the mic directly
into your computer’s USB port. Examples are the Samson C01U, Blue
Snowball, Rode Podcaster USB microphone; and MXL USB.009, 990
USB, and 990 USB stereo mics. The MXL MicMate plugs into any pro
mic and converts its output to USB. Phantom power is included. Its digital-audio format is 16-bit, 44.1 or 48 kHz. The Centrance MicPort Pro is
similar but records up to 24-bit/96 kHz audio.
Using a mouse can be fatiguing and can lead to repetitive stress
syndrome. An alternative to the mouse is a control surface or controller. It looks like a mixer with faders, but it adjusts the virtual
controls you see on the computer screen. That way you can use your
computer for recording, and control the software with knobs
and faders instead of a mouse. Two controllers with one fader are
the Frontier Design AlphaTrack ($200, Figure 4.8) and PreSonus
FaderPort ($150). Multi-fader controllers are made by such companies as Mackie, TASCAM, M-Audio, Digidesign, and Alesis ($350 to
$1300).
45
46
CHAPTER 4 Equipping Your Studio
FIGURE 4.8
Frontier Design AlphaTrack, an example of a controller for recording software.
KEYBOARD WORKSTATION
Here’s another device to record music. This is a synthesizer/sampler with a piano-style keyboard and a built-in sequencer (MIDI
recorder) and effects. Examples include Alesis Fusion, Promega 2 and
Andromeda; Korg K2600X; Yamaha Motif, Roland Fantom-X, Juno-G
and GW-7. The workstation lets you create a tune with several instrument sounds. Then you record the audio from the keyboard into a
computer. If you want to add a vocal, use MIDI/audio recording software. An arranger keyboard workstation automatically creates backing tracks (drums, bass, and chords) by following your left hand’s
notes and right-hand’s melodies. Examples include Yamaha Tyros2
and PSR-S900.
A similar device is a beat box or groove box, which is a sample player
with pads that you tap to generate sounds and grooves. It lets you
assemble a stereo music track from drum and synth samples. Copy
Equipping Your Studio CHAPTER 4
the music from the beat box into your computer and add vocals.
Examples include Akai MPC series and MPD24, Roland sampling
workstations, Zoom RT-223 and Streetboxx SB-246, Korg Electribe
series.
LEGACY RECORDING DEVICES
Some recording equipment has become almost obsolete, but you
may need to work with it if a client brings older gear into your studio to transfer audio tracks. Some of these formats are listed below.
Analog tape recorder: This records 2 to 24 tracks of analog audio
on reels of magnetic tape from 1/4 to 2 inches wide. Compared
to digital recorders, analog recorders have more noise, distortion, frequency-response errors, and unstable pitch. Their electronics and tape heads need frequent alignment. Still, many
people love the sound of them. Some have mic inputs; all have
analog line inputs and line outputs.
DAT recorder: This records 2 tracks of audio digitally on a small
DAT-tape, a cassette about half the size of a standard analog
cassette. The DAT also writes absolute time (hours, minutes,
and seconds) on tape as it records. Inputs are analog mic/line
level and digital; outputs are analog line level and digital.
Modular Digital Multitrack (MDM): This records 8 digital tracks
on a videocassette, using a rotating drum like a DAT recorder.
Two popular models are the Alesis ADAT which records on
S-VHS tape, and the TASCAM DA-88, DA-78 or DA-38 which
record on Hi-8 mm tape. ADAT records up to 40 minutes on a
single tape; DA-88 records up to 1 hour 48 minutes. With both
types, you can sync several 8-track units with a cable to add
more tracks. Both have analog line inputs and outputs. ADAT
has Lightpipe-format digital outputs; TASCAM models have
TDIF-format digital outputs.
Microphone
So far we covered recording devices. Let’s move on to other studio
equipment.
A microphone converts the sound of a voice or instrument into an
electrical signal that can be recorded. Microphone sound quality varies
47
48
CHAPTER 4 Equipping Your Studio
widely, so be sure to get one or more good mics costing at least $100
each. Your ears should tell you if a mic’s fidelity is adequate for your
purpose. Some people are happy to get any sound recorded; others settle for nothing less than professional sound quality.
Probably the most useful mic types for home recording are the cardioid condenser mic and cardioid dynamic mic. The cardioid pickup
pattern helps reject room acoustics for a tighter sound. The condenser
type is commonly used on cymbals, acoustic instruments, and studio
vocals; dynamics are used typically on drums and guitar amps. You’ll
also need at least one mic cable, mic stand, and boom costing about
$25 each. For more information on microphones, see Chapter 6.
Do you want to record solo instruments or classical musical ensembles in stereo with two mics at a distance? You’ll need either a stereo mic or a matched pair of high-quality condenser or ribbon mics
of the same model number, plus a stereo mic-stand adapter. See
Chapter 18 for details.
Phantom-Power Supply
A phantom-power supply powers the circuits in condenser mics. It
uses the same cable that carries the mic’s audio signal. You don’t
need the supply if your condenser mic has a battery, or if your
mixer or audio interface supplies phantom power at its mic connectors (most do). Some makers of phantom supplies are ART, AudioTechnica, Rolls, and Whirlwind. Prices range from $45 to $139.
Mic Preamp
This device amplifies a mic signal up to a higher voltage called “line
level,” which is required by mixers and recorders. A stand-alone
mic preamp provides a slightly cleaner, smoother, or more colorful
sound than a mic preamp built into a mixer or audio interface, but
costs much more. A 2-channel preamp costs $120 to $2000, while
8-channel preamps range from $600 to $6000. Studios on a budget
can do without a preamp.
Some manufacturers are Manley, True Systems, Focusrite, Universal
Audio, GLM, Chandler, A Designs, Millennia, Avalon, Great River,
John Hardy, Benchmark, AEA, Apogee, Vintech, Grace, Groove Tubes,
Presonus, Mackie, Summit Audio, Studio Projects, Blue, Samson,
dbx, Aphex, and M-Audio.
Equipping Your Studio CHAPTER 4
INSTRUMENT
AMP
GND
LIFT
MIC-LEVEL
OUTPUT
FIGURE 4.9
A direct box.
Direct Box (DI)
A direct box (Figure 4.9) is used to connect an electric instrument
(guitar, bass, synth) to an XLR-type mic input of a mixer, recorder–
mixer, or audio interface. It lets you record electric instruments
directly into your mixer without a microphone. A direct box picks
up a very clean sound, which may be undesirable for electric guitar.
If you want to pick up the distortion of the guitar amp, use a microphone instead. Or use a guitar-amp modeling device or modeling
plug-in.
Some recorder–mixers and audio interfaces have “high-impedance”
or “instrument” inputs meant for electric guitars and synthesizers.
In this case, simply use a short guitar cord between the instrument
and the mixer high-impedance input. If the cable run is over 15 feet,
you’ll get higher quality with a direct box plugged into an XLR mic
input.
Direct boxes sell for $50 to $700. Some manufacturers are Radial,
Countryman, Whirlwind, Manley, Groove Tubes, Pro Co, BSS, Tapco,
Aphex, and ART.
Monitor System
Another important part of your studio is the monitor system: a pair
of quality headphones or loudspeakers. The monitor system lets you
hear what you’re recording and mixing—what you’re doing to the
recorded sound. The sound you hear over the monitors is approximately what the final listener will hear.
49
50
CHAPTER 4 Equipping Your Studio
Very good headphones are available for $100 and up from Sony,
AKG, Sennheiser, Beyerdynamic, Audio Technica, Ultrasone, Shure,
and others. A headphone amp can power several headphones at
once, and some units let the musician adjust the headphone mix.
If you’re recording only yourself, one set of headphones is enough.
But if you’re recording another musician, you both need headphones.
Many recorder–mixers have two headphone jacks for this purpose.
If you want to record or overdub several people at once, you need
headphones for all of them. For example, if you’re overdubbing
three harmony vocalists, each one needs headphones to hear previously recorded tracks to sing with. To connect all these headphones,
you could build a headphone junction box: an aluminum or plastic
box that contains several headphone jacks. These are wired to a cable
coming from your mixer’s headphone jack. Or you could use a splitter cable, which makes two jacks out of one. Another option is to
purchase an affordable Rolls or Behringer headphone amplifier. Feed
the amp from aux sends or the stereo mix.
Powered speakers with built-in amplifiers provide accurate, high-fidelity
monitoring. Described in Chapter 5, these Nearfield studio monitors
are small, bookshelf-type speakers that are placed about 3 feet apart
and 3 feet from you as you sit at your mixer. Some monitor companies
include Genelec, Dynaudio, JBL, Event, Quested, Focal, Lipinski, KRK,
Blue Sky, Yamaha, Adam, Mackie, Alesis, Tannoy, M-audio, Behringer,
Edirol, and Samson. Prices range from $200 to $3860 a pair.
Effects
A recording without effects sounds rather dead and plain. Effects such
as reverberation, echo, and chorus can add sonic excitement to a recording. They are produced by devices called signal processors (see Figure
4.10) or by plug-ins, which are software effects used in a computer
recording program. See Chapter 10 for more information on effects.
Although effects are built into most recording programs and recorder–
mixers, most analog mixers require external effects units. On the mixer
is a set of connectors (labeled “aux send” and “aux return”) for hooking
up an external effects unit, such as a reverb or delay device. A unit with
one effects send lets you add one type of effect; a unit with two effects
sends lets you add two different effects to create more sonic interest.
Equipping Your Studio CHAPTER 4
FIGURE 4.10
An effects unit.
Miscellaneous Equipment
Other equipment for your home studio includes audio cables, USB or
FireWire cables, power outlet strips, lighting, tables or studio furniture,
mic pop filters, masking tape and a pen to label inputs and cables,
contact cleaning fluid, MIDI equipment stands, music stands, session
forms, connector adapters, pen and paper, a flashlight, and so on.
MIDI Studio Equipment
MIDI is covered in detail in Chapter 16. It describes the components
in a typical MIDI studio, such as shown in Figure 4.11. Also see
Chapter 2 under the heading MIDI Sequencing.
We’ve looked at several types of recording setups. All can help you
create quality demos. As you go higher in price you get more features and better sound. For example, if you want to record an entire
band playing all at once, with each instrument having its own mic,
you’ll need a system with more microphones, more tracks, and more
headphones.
As we’ve seen, putting together a home studio or project studio needn’t
cost much. As technology develops, better equipment is available at
lower prices. That dream of owning your own studio is within reach.
SETTING UP YOUR STUDIO
Once you have your equipment, you need to connect it together with
cables, and possibly install equipment racks and acoustic treatment.
Let’s consider each step.
Cables
Cables carry electric signals from one audio component to another.
They are usually made of one or two insulated conductors (wires)
surrounded by a fine-wire mesh shield that reduces hum. Outside
51
52
CHAPTER 4 Equipping Your Studio
MIDI / AUDIO
RECORDING SOFTWARE
multitrack audio recorder
sequencer
soft synths
sample players
AUDIO INTERFACE
WITH MIDI I/O
Computer audio out
to mixer or
powered speakers
MIDI
out
COMPUTER
MIDI
in
MIDI
SPLITTER
MIDI in
Optional sound module
audio out to mixer or
powered speakers
SOUND MODULE (optional)
MIDI in
MIDI out
Optional synth audio
out to mixer or
powered speakers
SYNTH OR KEYBOARD MIDI CONTROLLER
FIGURE 4.11
One type of MIDI studio.
the shield is a plastic or rubber insulating jacket. On both ends of
the cable are various types of connectors.
Cables are either balanced or unbalanced. A balanced line is a cable
that uses two wires (conductors) to carry the signal, surrounded by a
shield (see Figure 4.12). Each wire has equal impedance to ground.
An unbalanced line has a single conductor surrounded by a shield
(see Figure 4.13). The conductor and shield carry the signal. A balanced line rejects hum better than an unbalanced line, but an unbalanced line less than 10 feet long usually provides adequate hum
rejection and costs less.
A cable carries one of these five signal levels or voltages:
1. Mic level: About 2 mV (0.002 volt) to about 1 V depending on
how loud the sound source is, and on how sensitive the mic is.
2. Instrument level: Typically 0.1 to 1 V for passive pickups; up to
1.75 V for active pickups.
3. Semi-pro or consumer line level: ⫺10 dBV (0.316 volt).
4. Pro line level: ⫹4 dBu (1.23 volts).
5. Speaker level (about 20 volts).
Equipping Your Studio CHAPTER 4
CABLE JACKET
INSULATION
HOT CONDUCTOR
COLD CONDUCTOR
SHIELD
FIGURE 4.12
A 2-conductor shielded, balanced line.
CABLE JACKET
CONDUCTOR
SHIELD
FIGURE 4.13
A 1-conductor shielded, unbalanced line.
Note that “instrument level” can overlap mic level and line level.
Some manufacturers use dBu instead of volts (for an explanation,
see Appendix A):
Mic level: About ⫺52 dBu to ⫹2.2 dBu.
Instrument level: ⫺17.7 dBu to ⫹7 dBu.
Semi-pro or consumer line level: ⫺7.8 dBu
Pro line level: ⫹4 dBu.
Equipment Connectors
Recording equipment also has balanced or unbalanced connectors
built into the chassis. Be sure your cable connectors match your
equipment connectors.
Balanced equipment connectors:
■
■
3-pin (XLR-type) connector—Figure 4.14
1/4-inch TRS (tip-ring-sleeve) phone jack—Figure 4.15
Unbalanced equipment connectors:
■
■
1/4-inch TS (tip-sleeve) phone jack—Figure 4.15
Phono jack (RCA connector)—Figure 4.16
A jack is a receptacle; a plug inserts into a jack.
53
54
CHAPTER 4 Equipping Your Studio
PUSH
FIGURE 4.14
A 3-pin XLR-type connector used in balanced equipment. Left: male output connector. Right: female input
connector.
FIGURE 4.15
A 1/4-inch phone jack used in balanced and unbalanced equipment.
FIGURE 4.16
A phono (RCA) jack used in unbalanced equipment.
Connectors are confusing because a single connector can have several
functions (usually not at the same time). Here are some examples:
■
■
■
■
XLR: balanced line input at ⫹4 dBu, balanced mic input at 2 mV to 1 V, or
balanced line output at ⫹4 dBu.
TRS (stereo 1/4-inch phone jack): balanced mic input, insert send/return
connectors (line-level), instrument input, or stereo headphones.
TS (mono 1/4-inch phone jack): unbalanced mic input, unbalanced line-level
input or output (⫹4 dBu or ⫺10 dBV), instrument input, or low-cost speaker
connector.
Combi connector: an XLR mic input plus a TRS input (instrument-level or
line-level ).
Equipping Your Studio CHAPTER 4
FIGURE 4.17
A mono (TS) 1/4-inch phone plug.
FIGURE 4.18
An RCA (phono) plug.
■
■
RCA (phono): home stereo line-level input or output at ⫺10 dBV, composite
video input/output, or SPDIF digital-audio input/output.
1/8-inch (3.5 mm) mini phone jack: headphone jack, low-cost stereo mic
input, line output in a portable recorder at ⫺10 dBV; or a computer sound
card’s line in, line out, or speaker out.
Equipment connectors are labeled according to their function. If
you see an XLR connector with the label “MIC,” you know it’s a balanced mic input. If it’s an 1/8-inch connector on a sound card, look
at the icon near the connector. It’s either a mic input, line input, line
output, or speaker output. You could download the manual for the
sound card, which should describe the function of each connector.
Cable Connectors
Several types of cable connectors are used in audio. Figure 4.17
shows a 1/4-inch mono phone plug (or TS phone plug), used with
cables for unbalanced microphones, synthesizers, and electric instruments. The tip terminal is soldered to the cable’s center conductor;
the sleeve terminal is soldered to the cable shield.
Figure 4.18 shows an RCA or phono plug, used to connect unbalanced line-level signals. The center pin is soldered to the cable’s center conductor; the cup terminal is soldered to the cable shield.
55
56
CHAPTER 4 Equipping Your Studio
Figure 4.19 shows a 3-pin pro audio connector (XLR-type). It is used
with cables for balanced mics and balanced recording equipment. The
female connector (with holes; Figure 4.19A) plugs into equipment outputs. The male connector (with pins; Figure 4.19B) plugs into equipment inputs. Pin 1 is soldered to the cable shield, pin 2 is soldered
to the “hot” red or white lead, and pin 3 is soldered to the remaining
lead. This wiring applies to both female and male connectors.
Figure 4.20 shows a stereo (TRS) phone plug, used with stereo headphones and with some balanced line-level cables. For headphones, the
tip terminal is soldered to the left-channel lead; the ring terminal is
soldered to the right-channel lead; and the sleeve terminal is soldered
to the common lead. For balanced line-level cables, the sleeve terminal is soldered to the shield; the tip terminal is soldered to the hot red
or white lead; and the ring terminal is soldered to the remaining lead.
Some mixers have insert jacks that are stereo phone jacks; each jack
accepts a stereo phone plug. Tip is the send signal to an audio device
input, ring is the return signal from the device output, and sleeve is
ground.
FIGURE 4.19
A 3-pin pro audio connector (XLR-type). (A) female; (B) male.
FIGURE 4.20
A stereo (TRS) phone plug.
Equipping Your Studio CHAPTER 4
Cable Types
Cables are also classified according to their function. In a studio, you’ll
use several types of cables: power, mic, MIDI, speaker, USB, FireWire,
S/PDIF, TASCAM TDIF, Alesis Lightpipe, guitar cords, and patch cords.
A power cable, such as an AC extension cord or a power cord on a
device, is made of three heavy-gauge wires surrounded by an insulating jacket. The wires are thick to handle high current without
overheating.
A mic cable is usually 2-conductor shielded. It has two wires to carry
the signal, surrounded by a fine-wire cylinder or shield that reduces
hum pickup. On one end of the cable is a connector that plugs
into the microphone, usually a female XLR-type. On the other end
is either a 1/4-inch phone plug or a male XLR-type connector that
plugs into your mixer or audio interface.
Rather than running several mic cables to your mixer or interface, you
might consider using a snake: a box with multiple mic connectors, all
wired to a thick multiconductor cable (Figure 4.21). A snake is especially convenient if you’re running long cables to recording equipment
from another room. It’s essential for most on-location recording.
STAGE BOX
CONNECTORS TO
MIC CABLES
SNAKE
CONNECTORS TO MIXER
FIGURE 4.21
A stage box and snake.
57
58
CHAPTER 4 Equipping Your Studio
Professional balanced equipment is interconnected with mic cable:
2-conductor shielded cable having a female XLR on one end and a
male XLR on the other. Professional patch bays (described in the
next section) use balanced cables with TRS phone plugs.
A MIDI cable uses a 5-pin DIN connector on each end of a 2-conductor
shielded cable. The cable connects MIDI OUT to MIDI IN, or MIDI
THRU to MIDI IN.
A speaker cable connects a power amp to each loudspeaker. Speaker
cables are normally made of lamp cord (zip cord). To avoid wasting
power, speaker cables should be as short as possible and should be
heavy gauge (between 12 and 16 gauge). Number 12 gauge is thicker
than 14; 14 is thicker than 16.
A USB cable or a FireWire cable connects a peripheral device (like
an audio interface) to a computer. USB and FireWire are covered in
detail in Chapter 13.
An S/PDIF cable transfers a digital signal from one device’s S/PDIF
output to another device’s S/PDIF input. It uses a shielded unbalanced cable (ideally a 75-ohm RG59 video cable) with an RCA plug
on each end.
A TASCAM TDIF cable is a multiconductor cable with a 25-pin
D-sub connector on both ends. It’s used to connect multiple digitalaudio signals from TASCAM multitrack recorders to digital mixers or
computer TDIF interfaces.
An Alesis Lightpipe cable is a fiber-optic cable with a TOSLINK
connector on both ends. This cable is used to connect 8 channels of
digital-audio signals from an Alesis multitrack recorder to a digital
mixer or computer Lightpipe interface.
A guitar cord is constructed of a 1-conductor shielded cable with a
1/4-inch phone plug on each end. It connects a guitar amp to an
electric instrument (electric guitar, electric bass, synthesizer, or an
acoustic guitar with a pickup). Also, it connects an electric instrument to a direct box or to an instrument input.
Patch cords connect your recorder–mixer to external devices: an
effects unit, a 2-track recorder, and a power amplifier. They also connect an analog mixer to the analog inputs and outputs of a multitrack
Equipping Your Studio CHAPTER 4
recorder, usually as a snake that combines several cables. An unbalanced patch cord is made of 1-conductor shielded cable with either
a 1/4-inch phone plug or a phono (RCA) connector on each end.
A stereo patch cord is two patch cords joined side by side.
Rack/Patch Bay
You might want to mount your signal processors in a rack, an enclosure made of wood or metal with rails of screw holes for mounting
equipment (Figure 4.22). You also might want to install a patch panel
or patch bay: a group of connectors that are wired to equipment inputs
and outputs. Using a patch bay and patch cords, you can change
equipment connections easily. You also can bypass or patch around
defective equipment. Note that a patch bay increases the chance of
hum pickup slightly because of the additional cables and connectors.
Racks and patch bays aren’t essential, but they are convenient.
FIGURE 4.22
A rack and patch panel.
59
60
CHAPTER 4 Equipping Your Studio
EFFECTS
SENDS
CONSOLE
BUSSES OUT
1
2
3
4
2
3
4
1
2
STEREO
REVERB
OUT
STEREO
DELAY
OUT
RECORDER
LINE OUTPUTS
1
2
1
2
1
1
2
1
2
1
2
3
4
2
3
4
CONSOLE
INSERT
SENDS
1
2
OUTPUT
NORMAL
INPUT
1
RECORDER
LINE INPUTS
EFFECTS 1 EFFECTS 2
REVERB
RETURN RETURN
IN
DELAY
IN
CONSOLE
LINE INPUTS
1
2
CONSOLE
INSERT
RETURNS
Figure 4.23
Some typical patch-bay assignments.
Figure 4.23 shows some typical patch-panel assignments.
Equipment Connections
The instruction manuals of your equipment tell how to connect each
component to the others. In general, use cables that are as short as
possible to reduce hum, but that are long enough to let you make
changes.
Be sure to label all your cables on both ends according to what they
plug into—for example, MIXER CH1 MONITOR OUT or ALESIS
3630 IN. If you change connections temporarily, or the cable
becomes unplugged, you’ll know where to plug it back in. A piece of
masking tape folded over the end of the cable makes a stay-put label.
Typically, you follow this procedure to connect equipment (see
Figure 4.24):
1. Plug the AC power cords of audio equipment and electric musical instruments into AC outlet strips fed from the same circuit
breaker. Make sure that the sum of the equipment current ratings doesn’t exceed the breaker’s amp rating for that outlet. Plug
the power amplifier or powered speakers into their own outlet
on the same breaker so that they receive plenty of current.
2. Connect mic cables to mics.
3. Connect mic cables either to the snake junction box, or
directly into mic inputs on a mixer, mic preamps, or audio
Equipping Your Studio CHAPTER 4
POWER AMPLIFIER
AND MONITOR SPEAKERS,
OR POWERED MONITORS
MICS / DI
MICS
IN
MONITOR
OUT
L
R
SNAKE
LINE
OUT
MIXER
HEADPHONES
L
R
INSERT
SEND/RETURNS
LINE
IN
LINE
OUT
LINE
IN
IN / OUT
COMPUTER
WITH AUDIO
INTERFACE
MULTITRACK
RECORDER
(OPTIONAL)
FIGURE 4.24
Typical connections for recording-studio equipment.
FIGURE 4.25
A female XLR to phone plug adapter.
interface. Plug the snake connectors into the mic inputs. If
your mixer has phone-jack mic inputs, you may need to use
an impedance-matching adapter (female XLR to phone)
between the mic cable and the mic input jack (Figure 4.25).
61
62
CHAPTER 4 Equipping Your Studio
4. Set the output volume of synthesizers and sound modules
about three-quarters of the way up. Connect their audio outputs to instrument or line inputs on your mixer or audio
interface. If this causes hum, use a direct box.
If you’re recording a guitar direct, connect it either to (1) an instrument input on your mixer or audio interface, or (2) a direct box.
Connect the XLR output of the direct box to a mic input on your
mixer or audio interface.
5. If the mixer is a stand-alone unit (not part of a recorder–
mixer), connect the mixer’s stereo outputs to the inputs of an
audio interface. If the mixer has a USB or FireWire connector,
connect that to the mating connector in your computer.
6. Connect the audio interface audio outputs to the mixer’s
2-track or tape inputs, or to a volume-control unit, or directly
to powered speakers.
7. Connect the mixer’s monitor outputs to the power-amp inputs.
Connect the power-amp outputs to loudspeakers. Or if you’re
using powered (active) monitors, connect the mixer monitor
outputs to the monitor-speaker inputs.
8. If the mixer doesn’t have internal effects, connect the mixer
aux-send connectors to effects inputs (not shown). Connect the
effects outputs to the mixer aux-return or bus-in connectors.
9. If you’re using a separate mixer and multitrack recorder, connect mixer bus 1 to recorder track 1 IN, connect bus 2 to track
2, and so on. Also connect the recorder’s track 1 OUT to the
mixer’s line input 1, connect the track 2 OUT to the mixer’s
line input 2, and so on. As an alternative, connect insert jacks
to multitrack inputs and outputs. At each insert plug, connect the tip (send) terminal to a track input; connect the ring
(return) terminal to the same track’s output.
10. If you have several headphones for musicians, connect the
cue output to a small amplifier to drive their headphones. Or
if the mixer’s headphone signal is powerful enough, connect
it to a box with several headphone jacks wired in parallel.
Semi-pro studio equipment with unbalanced connectors (usually
phone or RCA) operates at a level called ⫺10 or ⫺10 dBV. Pro studio equipment with balanced connectors (XLR or TRS) works at ⫹4
or ⫹4 dBu. Check your equipment manuals to determine their input
and output levels. When you connect devices that run at different
Equipping Your Studio CHAPTER 4
POWER AMPLIFIER
AND MONITOR SPEAKERS,
OR POWERED MONITORS
MICS / DI
MICS
IN
SNAKE
HEADPHONES
MONITOR
OUT
AUDIO INTERFACE,
CONTROLLER SURFACE,
or
USB/FIREWIRE MIXER
PCI, USB or
FIREWIRE
COMPUTER
FIGURE 4.26
Typical connections in a DAW recording studio.
levels, set the ⫹4/⫺10 switch on each unit to match the levels. If there
is no such switch on either device, connect between them a ⫹4/⫺10
converter box such as the Whirlwind LM2U line-level converter
(www.whirlwindusa.com). Or try the cables shown in Appendix A
(dB or Not dB), Figures A.3 and A.4.
Figure 4.26 shows typical connections in a DAW recording studio
that uses an audio interface, controller surface, or USB/FireWire
mixer (described in Chapter 13). Figure 4.27 shows a typical layout
for a DAW recording studio.
A recorder–mixer studio can be quite simple. Plug mics into mic
inputs, plug headphones into headphone jacks, and plug powered
speakers into the mixer monitor outputs.
HUM PREVENTION
You patch in a piece of audio equipment, and there it is—HUM! It’s
a low-pitched tone or buzz. This annoying sound is a tone at 60 Hz
(50 Hz in Europe) and multiples of that frequency.
Hum is caused mainly by
1. Cables picking up magnetic and electrostatic hum fields radiated by power wiring—especially if the cable shield connection
is broken.
2. Ground loops. A ground loop is a closed loop of ground wires.
It’s made of two separated pieces of audio equipment each
63
64
CHAPTER 4 Equipping Your Studio
LEFT
SPEAKER
RIGHT
SPEAKER
COMPUTER
MONITOR
HEADPHONES
AUDIO INTERFACE,
CONTROLLER SURFACE,
or USB/FIREWIRE MIXER
PCI, USB or
FIREWIRE
TO MIC INPUTS
COMPUTER
KEYBOARD
COMPUTER
SNAKE
MICS
MULTITRACK
RECORDER
(OPTIONAL)
FIGURE 4.27
Typical layout of a DAW recording studio.
connected to power ground through a 3-prong power cord,
and also connected to each other through a cable shield. The
ground voltage may be slightly different at each piece of equipment, so a 50- or 60-Hz hum signal flows between the components along the cable shield.
These are the most important points to remember about hum
prevention:
■
■
■
■
To prevent ground loops, plug all equipment into outlet strips
powered by the same AC outlet.
Do not use an AC (electrical) 3-to-2 adapter to disconnect the
power ground—this creates a safety hazard.
Some power amps create hum if they don’t get enough AC current. So connect the power amp (or powered speakers) AC plug
to its own wall outlet socket—the same outlet that feeds the
outlet strips for the recording equipment.
If possible, use balanced cables going into balanced equipment.
Balanced cables have XLR or TRS connectors and two conductors
Equipping Your Studio CHAPTER 4
■
■
surrounded by a shield. At both ends of the cable, connect the
shield to a screw in the chassis, not to XLR pin 1. Or use audio
gear whose XLR connectors are wired with pin 1 to chassis
ground, not to signal ground.
Transformer-isolate unbalanced connections. If that isn’t an
option, use the cable assemblies shown in Figures A.3 and A.4
in Appendix A.
Don’t use conventional SCR dimmers to change the studio
lighting levels. Use Luxtrol® variable-transformer dimmers or
multi-way incandescent bulbs instead.
Even if your system is wired properly, a hum or buzz may appear
when you make a connection. Follow these tips to stop the hum:
■
■
■
■
■
■
■
■
■
■
■
■
■
If the hum is coming from a direct box, flip its ground-lift switch.
Check cables and connectors for broken leads and shields.
Unplug all equipment from each other. Start by listening just
to the powered monitor speakers. Connect a component to the
system one at a time, and see when the hum starts.
Remove audio cables from your devices and monitor each
device by itself. It may be defective.
Lower the volume on your power amp (or powered speakers),
and feed them a higher level signal.
Use a direct box instead of a guitar cord between instrument
and mixer.
To stop a ground loop when connecting two devices, connect
between them a 1:1 isolation transformer, direct box or hum
eliminator (such as Jensen CI-2RR, Behringer HD400, Rolls HE18
or Ebtech He2PKG). See Figures A.3 and A.4 in Appendix A.
Make sure that the snake box isn’t touching metal.
To prevent accidental ground loops, don’t connect XLR pin 1 to
the connector shell except for permanent connections to equipment inputs and outputs.
Try another mic.
If you hear a hum or buzz from an electric guitar, have the player
move to a different location or aim in a different direction. You
might also attach a wire between the player’s body and the guitar strings near the tailpiece to ground the player’s body.
Turn down the high-frequency EQ on a buzzing bass-guitar track.
To reduce buzzing between notes on an electric-guitar track,
apply a noise gate.
65
66
CHAPTER 4 Equipping Your Studio
■
■
Route mic cables and patch cords away from power cords;
separate them vertically where they cross. Also keep recording
equipment and cables away from computer monitors, power
amplifiers, and power transformers.
See Rane’s excellent article on sound system interconnections at
www.rane.com.
By following all these tips, you should be able to connect audio
equipment without introducing any hum. Good luck!
REDUCING RADIO FREQUENCY
INTERFERENCE (RFI)
RFI is heard as buzzing, clicks, radio programs, or “hash” in the audio
signal. It’s caused by CB transmitters, computers, lightning, radar, radio
and TV transmitters, industrial machines, auto ignitions, stage lighting,
and other sources. Many of the following techniques are the same used
to reduce hum from other sources. To reduce RFI do the following:
■
■
■
■
■
■
If you think that a speaker cable, mic cable, or patch cord is
picking up RFI, wrap the cable several times around an RFI
choke (available at Radio Shack). Put the choke near the device
that is receiving audio.
Install high-quality RFI filters in the AC power outlets. The
cheap types available from local electronics shops are generally
ineffective.
If the shield is disconnected in a balanced cable connector, solder it to pin 1.
If a mic is picking up RFI, solder a 0.047 microfarad capacitor
between pin 1 and 2 and between pin 1 and 3 in the female
XLR connector of the mic cable. Caution: this might cause highfrequency distortion with some mics.
At equipment XLR mic inputs, use a cable connector with a tab
that is connected to the metal shell. Use a wide strip of metal
braid to connect this tab to pin 1. Make sure the connector
makes metal-to-metal contact with the chassis.
Periodically clean connector contacts with Caig Labs DeoxIT, or
at least unplug and plug them in several times.
This chapter briefly covered the equipment and connectors for a
recording studio. The rest of this book explains each piece of equipment in detail and tells how to use it for best results.
CHAPTER 5
Monitoring
One of the most exciting moments in recording comes when the finished mix is played over the studio monitor speakers. The sound is
so clear you can hear every detail, and so powerful you can feel the
deep bass throbbing in your chest.
You use the monitor system to listen to the output signals of the console, audio interface, or recorders. It consists of the console monitor
mixer, the power amplifiers, loudspeakers, and the listening room.
The power amplifier boosts the electrical power of the console signal
to a sufficient level to drive a loudspeaker. The speaker converts the
electrical signal into sound, and the listening-room acoustics affect
the sound from the speaker.
A quality monitor system is a must if you want your mixes to sound
good. The power amp and speakers tell you what you’re doing to the
recorded sound. According to what you hear, you adjust the mix and
judge your mic techniques. Clearly, the monitor system affects the
settings of many controls on your mixer, as well as your mic selection
and placement. And all those settings affect the sound you’re recording. So, using inadequate monitors can result in a poor-sounding
product coming out of your studio.
It’s important to use accurate speakers that have a flat frequency
response. If your monitors are weak in the bass, you will tend to
boost the bass in the mix until it sounds right over those monitors.
But when that mix is played over speakers with a flatter response, it
will sound too bassy because you boosted the bass on your mixer.
So, using monitors with weak bass results in bassy recordings; using
monitors with exaggerated treble results in dull recordings, and so
67
68
CHAPTER 5 Monitoring
on. In general, colorations in the monitors will be inverted in your
mixdown recording.
That’s why it’s so important to use an accurate monitor system—one
with a wide, smooth frequency response. Such a system lets you hear
exactly what you recorded.
SPEAKER REQUIREMENTS
The requirements for an accurate studio monitor include:
■
■
■
■
■
■
Wide, smooth frequency response. To ensure accurate tonal
reproduction, the on-axis response of the direct sound should be
⫾4 dB or less from at least 40 Hz to 15 kHz. The low-frequency
response of a small monitor speaker should extend to at least
70 Hz.
Uniform off-axis response. The high-frequency output of a
speaker tends to diminish off-axis. Ideally the response at 30
degrees off-axis should be only a few decibels down from the
response on-axis. That way, a producer and engineer sitting side
by side will hear the same amount of treble. Also, the treble
won’t change as the engineer moves around at the console.
Good transient response. This is the ability of the speaker to
accurately follow the attack and decay of musical sounds. If
a speaker has good transient response, the bass guitar sounds
tight, not boomy, and drum hits have sharp impact. Some
speakers are designed so that the woofer and tweeter signals are
aligned in time. This aids transient response.
Clarity and detail. You should be able to hear small differences
in the sonic character of instruments, and to sort them out in a
complex musical passage.
Low distortion. Low distortion is necessary because it lets you
listen to the speaker for a long time without your ears hurting.
A good spec might be: Total harmonic distortion under 3%
from 40 Hz to 20 kHz at 90 dBSPL (sound pressure level).
Sensitivity. This is the sound pressure level a speaker produces
at 1 meter (m) when driven with 1 watt (W) of pink noise. Pink
noise is random noise with equal energy per octave. This noise
is either band-limited to the range of the speaker or is a onethird-octave band centered at 1 kHz. Sensitivity is measured in
dB/W/m (dB sound pressure level per 1 W at 1 m). A spec of
Monitoring CHAPTER 5
■
93 dB/W/m is considered high; 85 dB/W/m is low. The higher
the sensitivity, the less amplifier power you need to get adequate loudness.
High output capability. This is the ability of a speaker to play
loudly without burning out. You often need to monitor at high
levels to hear quiet details in the music. Plus, when you record
musicians who play loudly in the studio, it can be a letdown for
them to hear a quiet playback. So you may need a maximum
output of 110 dBSPL.
This formula calculates the maximum output of a speaker (how loud
it can play):
dBSPL ⫽ 10 log(P) ⫹ S
where dBSPL is the sound pressure level at 1 m, P is the continuous
power rating of the speaker in Watts, and S is the sensitivity rating in
dB/W/m.
For example, if a speaker is rated at 100 W maximum continuous
power, and its sensitivity is 94 dBSPL/W/m, its maximum output SPL
is 10log(100) ⫹ 94 ⫽ 114dBSPL (at 1 m from the speaker). The level
at 2 m will be about 4 to 6 dB less.
NEARFIELD™ MONITORS
Many professional recording studios use large monitor speakers that
have deep bass. However, they are expensive, heavy, and difficult to
install, and they are affected by the acoustics of the control room.
All studios, large or small, need a pair of Nearfield monitor speakers
(Figure 5.1). A Nearfield monitor is a small, wide-range speaker typically using a cone woofer and dome-shaped tweeter. You place a pair
of them about 3 or 4 feet apart, on stands just behind the console,
about 3 or 4 feet from you. Nearfields are far more popular than
large wall-mounted speakers.
This technique, developed by audio consultant Ed Long, is called
Nearfield monitoring. Because the speakers are close to your ears,
you hear mainly the direct sound of the speakers and tend to ignore
the room acoustics. Plus, Nearfield monitors sound very clear and
provide sharp stereo imaging. Some units have bass or treble tone
69
70
CHAPTER 5 Monitoring
FIGURE 5.1
A Nearfield monitor speaker.
controls built in to compensate for the effects of speaker placement
and room surfaces.
Nearfield monitors have enough bass to sound full when placed far
from walls. Although most Nearfields lack deep bass, they can be
supplemented with a subwoofer to reproduce the complete audio
spectrum. Or you can check the mix occasionally with headphones
that have deep bass.
Some Nearfields are in a satellite-subwoofer format. The two satellite
speakers are small units, typically including a 4-inch woofer and 3/4inch dome tweeter. The satellites are too small to produce deep bass,
but that is handled by the subwoofer—a single cabinet with one or
two large woofer cones. Typically, the subwoofer (sub) produces frequencies from 100 Hz down to 40 Hz or below. Because we don’t
localize sounds below about 100 Hz, all of the sound seems to come
from the satellite speakers. The sub-satellite system is more complicated to set up than two larger speakers, but offers deeper bass.
Monitoring CHAPTER 5
FIGURE 5.2
Crown D-75A, an example of a power amplifier.
POWERED (ACTIVE) MONITORS
Most monitors have a power amplifier built in. You feed them a linelevel signal (labeled MONITOR OUT) from your mixing console or
audio interface. Most powered monitors are bi-amplified: they have
one amplifier for the woofer and another for the tweeter. The advantages of bi-amplification include:
■
■
■
■
■
■
Distortion frequencies caused by clipping the woofer power
amplifier won’t reach the tweeter, so there is less likelihood of
tweeter burnout if the amplifier clips. In addition, clipping distortion in the woofer amplifier is made less audible.
Intermodulation distortion is reduced.
Peak power output is greater than that of a single amplifier of
equivalent power.
Direct coupling of amplifiers to speakers improves transient
response—especially at low frequencies.
Bi-amping reduces the inductive and capacitive loading of the
power amplifier.
The full power of the tweeter amp is available regardless of the
power required by the woofer amp.
THE POWER AMPLIFIER
If your monitor speakers aren’t powered, you need a power amplifier
(Figure 5.2). It boosts your mixer’s line-level signal to a higher power
to drive the speakers.
How many watts of power do you need? The monitor speaker’s
data sheet gives this information. Look for the specification called
“Recommended amplifier power.” A power amp of 50 W per channel
continuous is about the minimum for Nearfield monitors; 150 W is
better. Too much power is better than too little, because an underpowered system is likely to clip or distort. This creates high frequencies that can damage tweeters.
71
72
CHAPTER 5 Monitoring
A good monitor power amp has distortion under 0.05% at full
power. It should have a high damping factor—at least 100—to keep
the bass tight. The amp should be reliable. Look for separate level
controls for left and right channels. The amplifier should have a clip
or peak light that flashes when the amp is distorting.
SPEAKER CABLES AND POLARITY
When you connect the power amp to the speakers, use good wiring
practice. Long or thin cables waste amplifier power by heating. So put
the power amp(s) close to the speakers and use short cables with thick
conductors—at least 16 gauge. The low resistance of these cables helps
the power amplifier to damp the speaker motion and tighten the bass.
If you wire the two speakers in opposite polarity, one speaker’s cone
moves out while the other speaker’s cone moves in. This causes
vague stereo imaging, weak bass, and a strange sense of pressure on
your ears. Be sure to wire the speakers in the same polarity as follows: In both channels, connect the amplifier positive (or red) terminal to the speaker positive (or red) terminal. Setting the correct
polarity is also called “speaker phasing.”
CONTROL-ROOM ACOUSTICS
The acoustics of the control room affect the sound of the speakers.
At your ears, the sound waves reflected from the room surfaces combine with the direct sound from the speakers. Reflections that arrive
within 20 to 65 milliseconds (msec) after the direct sound blend
with the direct sound and affect the tonal balance you hear.
Chapter 3, under the heading Controlling Room Problems with
Acoustic Treatments, described how to treat the acoustics of studios
and control rooms to reduce excessive reverb, echo, standing waves,
and noise.
Using Nearfield monitors makes the room acoustics less important, but it still helps to put absorbent material on the wall behind
the monitors. This treatment improves the monitors’ sound. Stereo
imaging and depth are greatly improved, the sound is clearer, and
the frequency response is flatter. The treatment will reduce boominess and ringing, and make transients sharper. Also, your recordings
will translate better to other speakers.
Monitoring CHAPTER 5
While sitting in front of your mixer, have someone slide a mirror
along the walls at eye height. Note the spots where you can see the
monitor speakers in the mirror, and put some absorbent material
there to absorb sound reflections. Repeat this procedure with the
mirror on the ceiling between the monitors and your mix position.
If your control room is separate from the studio, the control room
should be built to keep out sound from the studio. You want to hear
only the sound from the monitors, not the live sound from the musicians. In a home studio, you might be able to achieve isolation simply by putting the control-room equipment in a room far removed
from the studio, with the doors closed.
A control room built next to the studio needs good isolation. Use
staggered-stud construction, seal holes between rooms, and install
a double-pane window (mounted in rubber) between the control
room and studio.
In some home or project studios, the control room is the same room
as the studio. Because no isolation is used, the cost of building the
studio is much less. You record while listening with headphones,
and do the critical monitoring with speakers during playback and
mixdown.
SPEAKER PLACEMENT
Once you have acquired the speakers and worked on the room
acoustics, you can install the speakers. Mount them at ear height so
the mixer doesn’t block their sound. To prevent sound reflections off
the mixing console, place the speakers on stands behind the console’s meter bridge, rather than putting them on top. For best stereo
imaging, align the speaker drivers vertically and mount the speakers
symmetrically with respect to the side walls. Place the two speakers as far apart as you’re sitting from them; aim them toward you,
and sit exactly between them (Figure 5.3). To get the smoothest lowfrequency response, put the speakers near the shorter wall, and sit
forward of the halfway point in the room.
Try to position the monitors several feet from the nearest wall. Wall
reflections can degrade the frequency response and stereo imaging. The closer to the wall the monitors are, the more bass you hear.
In small rooms you might have to place the monitors against the
73
74
CHAPTER 5 Monitoring
SPKR
SPKR
D
D
D
LISTENER
FIGURE 5.3
The recommended speaker/listener relationship for best stereo imaging.
wall, which will exaggerate the bass. But some monitors have a lowfrequency attenuation switch to compensate.
USING THE MONITORS
You’ve treated the room acoustics, and you’ve connected and placed
the speakers as described earlier. Now it’s time to adjust the stereo
balance.
1. Play a mono musical signal into an input channel on your
mixer, and assign it to the stereo output channels 1 and 2.
2. Adjust the input channel’s pan pot so that the signal reads the
same on the stereo output channel 1 and 2 meters.
3. Place the two speakers the same distance from you.
4. Sit at the mixer exactly midway between the speakers. If you
sit off-center, you will hear the image shifted toward one side.
Listen to the image of the sound between the speaker pair.
You should localize it midway between the monitors; that is,
straight ahead.
5. If necessary, center the image by adjusting the left or right volume control on your power amp or powered monitors.
When you do a mixdown, try to keep the listening level around
85 dBSPL—a fairly loud home listening level. As discovered by
Fletcher and Munson, we hear less bass in a program that is played
quietly than in the same program played loudly. If you mix a program while monitoring at, say, 100 dBSPL, the same program will
Monitoring CHAPTER 5
sound weak in the bass when heard at a lower listening level—which
is likely in the home. So, programs meant to be heard at 85 dBSPL
should be mixed and monitored at that level.
Loud monitoring also exaggerates the frequencies around 4 kHz.
A recording mixed loud may sound punchy, but the same recording
heard at a low volume will sound dull and lifeless.
Here’s another reason to avoid extreme monitor levels: Loud sustained sound can damage your hearing or cause temporary hearing
loss at certain frequencies. If you must do a loud playback for the
musicians (who are used to high SPLs in the studio), protect your
ears by wearing earplugs or leaving the room.
You can get a low-cost sound level meter from Radio Shack. Play a
musical program at 0 VU or 0 dB on the mixer meters and adjust the
monitor level to obtain an average reading of 85 dBSPL on the sound
level meter. Mark the monitor-level setting.
Before doing a mix, you may want to play some familiar commercial
CDs over your monitors to remind yourself what a good tonal balance sounds like. Listen to the amount of bass, midrange and treble, and try to match those in your mixes. But listen to several CDs
because they vary.
While mixing, monitor the program alternately in stereo and mono
to make sure there are no out-of-phase signals that cancel certain
frequencies in mono. Also beware of center-channel buildup: instruments or vocals that are panned to center in the stereo mix sound
3 dB louder when monitored in mono than they do in stereo. That
is, the balance changes in mono—the center instruments are a little
too loud. To prevent this, don’t pan tracks hard left and hard right.
Bring in the side images a little so they will be louder in mono.
You’ll mix the tracks to sound good on your accurate monitors.
But also check the mix on small inexpensive speakers to see whether
anything is missing or whether the mix changes drastically. Make
sure that bass instruments are recorded with enough edge or harmonics to be audible on the smaller speakers. It’s a good idea to
make a CD copy of the mix for listening in a car, boom box, or compact stereo.
75
76
CHAPTER 5 Monitoring
HEADPHONES
Compared to speakers, headphones have several advantages:
■
■
■
■
■
■
They cost much less.
There is no coloration from room acoustics.
The tone quality is the same in different environments.
They are convenient for on-location monitoring.
It’s easy to hear small changes in the mix.
Transients are sharper due to the absence of room reflections.
Headphones have several disadvantages:
■
■
■
■
■
■
■
They become uncomfortable after long listening sessions.
Cheap headphones have inaccurate tone quality.
Headphones don’t project bass notes through your body.
The bass response varies due to changing headphone pressure.
The sound is in your head rather than out front.
You hear no room reverberation, so you may add too much or
too little artificial reverb.
It’s difficult to judge the stereo spread. Over headphones,
panned signals tend to sound closer to center than the same
signals heard over speakers. The same is true of stereo recordings made with a coincident pair of mics.
Because speakers sound different from headphones, it’s best to
do mixes over speakers. But check your mixes on headphones too
because so many listeners use them with MP3 players.
If your monitor speakers are in the same room as your microphones,
the mics pick up the sound of the speakers. This causes feedback or a
muddy sound. In this case you must monitor only with headphones
while recording or overdubbing, then monitor with speakers during
playback or mixdown.
If you monitor with headphones as the musicians are playing, external sounds leak through the headphones and mask the sounds
you’re trying to monitor. That makes it hard to judge the sound
quality you’re picking up. Check it during playback. Closed-cup
headphones or in-the-ear earphones provide the best isolation from
outside sounds.
Monitoring CHAPTER 5
THE CUE SYSTEM
The cue system is a monitor system for musicians to use as they’re
recording. It includes some of the aux knobs in your mixer, a small
power amplifier or headphone amp, a headphone connector box,
and headphones. Musicians often can’t hear each other well in the
studio, but they can listen over headphones to hear each other in
a good balance. Also, they can listen to previously recorded tracks
while overdubbing.
Headphones for a cue system should be durable and comfortable.
They should be closed-cup to avoid leakage into microphones. This
is an ideal situation; open-air phones may work well enough. Also,
the cue “phones” should have a smooth response to reduce listening fatigue, and should play loud without burning out. Make sure
they are all the same model so each musician hears the same thing.
A built-in volume control is convenient.
A suggested cue system is shown in Figure 5.4. Connect a power amp
to an aux-send or monitor output of your mixer. The amp drives several resistor-isolated headphones, which are in parallel; that is, all
100-OHM
5-WATT
RESISTORS
(one per channel)
SMALL
POWER
AMP
MIXER
CUE OUT
OR
MONITOR OUT
STEREO
PHONE
JACKS
FIGURE 5.4
A cue system.
HEADPHONES
77
78
CHAPTER 5 Monitoring
the tip (left-channel) terminals on the jacks are wired together, and
all the ring (right-channel) terminals on the jacks are wired together.
If your mixer has a strong signal at its headphone jack, you can get
by with four headphone jacks in a small metal box wired in parallel
and connected to the mixer’s headphone jack.
Although some consoles can provide several independent cue mixes,
the ideal situation is to set up individual cue mixers near each musician. Then they can set their own cue mix and listening level. The
inputs of these mixers are fed from the console output busses.
Suppose a vocalist sings into a microphone and hears that mic’s
signal over the cue headphones. If the singer’s voice and the headphone’s sound are opposite in polarity, the voice partially cancels or
sounds funny in the headphones. Make sure that the voice and headphones are the same polarity.
Here’s how. While talking into a mic and listening to it on headphones, reverse the ground and signal leads to the headphones connector. The position that gives the fullest, most solid sound in the
headphones is correct.
All the headphones in your studio should be the same model, so
that everyone will hear with correct polarity.
CONCLUSION
Ultimately, what you hear from the monitors influences your recording techniques and affects the quality of your recordings. So take
the time to plan and adjust the control-room acoustics. Choose and
place the speakers carefully. Monitor at proper levels and listen on
several systems. You’ll be rewarded with a monitor system you can
trust.
CHAPTER 6
Microphones
What microphone is best for recording an orchestra? What’s a good
snare mic? Should the microphone be a condenser or dynamic,
omni or cardioid?
You can answer these questions more easily once you know the types
of microphones and understand their specs. First, it always pays to
get a high-quality microphone. The mic is a source of your recorded
signal. If that signal is noisy, distorted, or tonally colored, you’ll be
stuck with those flaws through the whole recording process. Better to
get it right up front.
Even if you have a MIDI studio and get your sounds from samples or
synthesizers, you still might need a good microphone for sampling,
or to record vocals, sax, acoustic guitar, and so on.
A microphone is a transducer—a device that changes one form of
energy into another. Specifically, a mic changes sound into an electrical signal. Your mixer amplifies and modifies this signal.
TRANSDUCER TYPES
Mics for recording can be grouped into three types depending on how
they convert sound to electricity: dynamic, ribbon, or condenser.
A dynamic mic capsule, or transducer, is shown in Figure 6.1. A coil
of wire attached to a diaphragm is suspended in a magnetic field.
When sound waves vibrate the diaphragm, the coil vibrates in
the magnetic field and generates an electrical signal similar to the
79
80
CHAPTER 6 Microphones
VOICE
COIL
DIAPHRAGM
SIGNAL
+
– +
MAGNET
DIAPHRAGM
MOVEMENT
–
MAGNET
FIGURE 6.1
A dynamic transducer.
POLE PIECE
MAGNET
RIBBON
FRONT
REAR
SIGNAL
FIGURE 6.2
A ribbon transducer.
incoming sound wave. Another name for a dynamic mic is movingcoil mic, but this term is seldom used.
In a ribbon mic capsule, a thin metal foil or ribbon is suspended in
a magnetic field (Figure 6.2). Sound waves vibrate the ribbon in the
field and generate an electrical signal.
A condenser or capacitor mic capsule has a conductive diaphragm
and a metal backplate placed very close together (Figure 6.3). They
are charged with static electricity to form two plates of a capacitor.
When sound waves strike the diaphragm, it vibrates. This varies the
spacing between the plates. In turn, this varies the capacitance and
generates a signal similar to the incoming sound wave. Because of
Microphones CHAPTER 6
SIGNAL
+
–
DIAPHRAGM
DIAPHRAGM
MOVEMENT
– +
BACKPLATE
INSULATING
SPACER
FIGURE 6.3
A condenser transducer.
its lower diaphragm mass and higher damping, a condenser mic
responds faster than a dynamic mic to rapidly changing sound waves
(transients).
Two types of condenser mic are true condenser and electret condenser. In a true condenser mic (externally biased mic), the diaphragm and backplate are charged with a voltage from a circuit built
into the mic. In an electret condenser mic, the diaphragm and backplate are charged by an electret material, which is in the diaphragm
or on the backplate. Electrets and true condensers can sound equally
good, although some engineers prefer true condensers, which tend
to cost more.
A condenser mic needs a power supply to operate, such as a battery or
phantom power supply. Phantom power is 12 to 48 volts DC applied
to pins 2 and 3 of the mic connector through two equal resistors.
The microphone receives phantom power and sends audio signals
on the same two conductors. Ground for the phantom power supply
is through the cable shield. Nearly all mixing consoles and audio
interfaces supply phantom power at their mic input connectors. You
simply plug the mic into the mixer to power it.
81
82
CHAPTER 6 Microphones
TYPICAL DYNAMIC VOCAL MIC
GRILLE
WIRES
MIC CAPSULE
CONNECTOR
SHOCK MOUNT
DIAPHRAGM
DIAPHRAGM
GRILLE
HOUSING
MAGNET
REAR PORT
HOUSING
CONNECTOR
CIRCUIT BOARD
TYPICAL CONDENSER INSTRUMENT MIC
FIGURE 6.4
Inside a typical dynamic vocal mic and condenser instrument mic.
Dynamics and ribbons need no power supply. You can plug these
types of mics into a phantom supply without damage because the
voice coil or ribbon isn’t connected to ground (unless they are accidentally shorted to the mic housing). Some newer ribbon mics have
a built-in preamp that is phantom powered.
Figure 6.4 shows a cutaway view of a typical dynamic vocal mic and
condenser instrument mic.
General Traits of Each Transducer Type
Condenser
■
■
■
■
■
■
Wide, smooth frequency response
Detailed sound, extended highs
Omni type has excellent low-frequency response
Transient attacks sound sharp and clear
Preferred for acoustic instruments, cymbals, studio vocals
Can be miniaturized
Microphones CHAPTER 6
Dynamic
■
■
■
■
■
■
Tends to have rougher response, but still quite usable
Rugged and reliable
Handles heat, cold, and high humidity
Handles high volume without distortion
Preferred for guitar amps and drums
If flat response, can take the “edge” off woodwinds and brass
Ribbon
■
■
■
Prized for its warm, smooth tone quality
Delicate
Complements digital recording
There are exceptions to the tendencies listed above. Some dynamics have a smooth, wide-range frequency response. Some condensers are rugged and handle high SPLs. It depends on the specs of the
particular mic. Audio clip 13 at www.elsevierdirect.com/companions/
9780240811444 demonstrates the sound of each transducer type.
POLAR PATTERN
Microphones also differ in the way they respond to sounds coming
from different directions. An omnidirectional microphone is equally
sensitive to sounds arriving from all directions. A unidirectional mic
is most sensitive to sound arriving from one direction—in front of
the mic—but softens sounds entering the sides or rear of the mic.
A bidirectional mic is most sensitive to sounds arriving from two
directions—in front of and behind the mic—but rejects sounds
entering the sides.
There are three types of unidirectional patterns: cardioid, supercardioid, and hypercardioid. A mic with a cardioid pattern is sensitive
to sounds arriving from a broad angle in front of the mic. It’s about
6 dB less sensitive at the sides, and about 15 to 25 dB less sensitive in
the rear. The supercardioid pattern is 8.7 dB less sensitive at the sides
and has two areas of least pickup at 125 degrees away from the front.
The hypercardioid pattern is 12 dB less sensitive at the sides and has
two areas of least pickup at 110 degrees away from the front.
To hear how a cardioid pickup pattern works, talk into a cardioid
mic from all sides while listening to its output. Your reproduced
83
84
CHAPTER 6 Microphones
voice is loudest when you talk into the front of the mic, and softest when you talk into the rear. Play audio clip 14 at www.elsevierdirect
.com/companions/9780240811444.
The super- and hypercardioid mics reject sound from the sides more
than the cardioid. They are more directional, but they pick up more
sound from the rear than the cardioid.
A microphone’s polar pattern is a graph of its sensitivity versus the
angle at which sound approaches the mic. The polar pattern is plotted on polar graph paper. Sensitivity is plotted as distance from the
origin. Figure 6.5 shows various polar patterns.
Traits of Each Polar Pattern
Omnidirectional
■
■
■
■
■
■
■
■
All-around pickup
Most pickup of room reverberation (play audio clip 14 at www
.elsevierdirect.com/companions/9780240811444)
Not much isolation unless you mike close
Low sensitivity to pops (explosive breath sounds)
Low handling noise
No up-close bass boost (proximity effect)
Extended low-frequency response in condenser mics—great for
pipe organ or bass drum in an orchestra or symphonic band
Lower cost in general
Unidirectional (cardioid, supercardioid, hypercardioid)
■
■
■
■
■
■
Selective pickup
Rejection of room acoustics, background noise, and leakage
Good isolation—good separation between tracks
Up-close bass boost (except in mics that have holes in the
handle)
Better gain-before-feedback in a sound-reinforcement system
Coincident or near-coincident stereo miking (explained in
Chapter 7)
Cardioid
■
■
■
Broad-angle pickup of sources in front of the mic
Maximum rejection of sound approaching the rear of the mic
Most popular pattern
Microphones CHAPTER 6
FRONT
0⬚
330⬚
FRONT
0⬚
30⬚
⫺5
330⬚
⫺5
⫺10
⫺15
300⬚
60⬚
⫺15
300⬚
⫺20
⫺20
⫺25
⫺25
270⬚
90⬚
240⬚
210⬚
90⬚
240⬚
120⬚
150⬚
150⬚
210⬚
180⬚
REAR
180⬚
REAR
OMNIDIRECTIONAL
CARDIOID
FRONT
0⬚
FRONT
0⬚
30⬚
⫺5
330⬚
⫺5
⫺10
⫺15
300⬚
60⬚
270⬚
120⬚
330⬚
30⬚
⫺10
30⬚
⫺10
60⬚
⫺15
300⬚
⫺20
60⬚
⫺20
⫺25
⫺25
90⬚
270⬚
270⬚
120⬚
240⬚
90⬚
240⬚
150⬚
210⬚
120⬚
150⬚
210⬚
180⬚
REAR
180⬚
REAR
SUPERCARDIOID
HYPERCARDIOID
FRONT
0⬚
330⬚
⫺5
30⬚
⫺10
⫺15
300⬚
60⬚
⫺20
⫺25
270⬚
90⬚
240⬚
120⬚
150⬚
210⬚
180⬚
REAR
BIDIRECTIONAL
FIGURE 6.5
Various polar patterns. Sensitivity is plotted vs. angle of sound incidence.
85
86
CHAPTER 6 Microphones
Supercardioid
■
■
■
Maximum difference between front hemisphere and rear hemisphere pickup (good for stage-floor miking)
More isolation than a cardioid
Less reverb pickup than a cardioid
Hypercardioid
■
■
Maximum side rejection in a unidirectional mic
Maximum isolation—maximum rejection of reverberation,
leakage, feedback, and background noise
Bidirectional
■
■
■
Front and rear pickup, with side sounds rejected (for acrosstable interviews or two-part vocal groups, for example)
Maximum isolation of an orchestral section when miked
overhead
Blumlein stereo miking (two bidirectional mics crossed at
90 degrees)
In a good mic, the polar pattern should be about the same from
200 Hz to 10 kHz. If not, you’ll hear off-axis coloration: the mic will
have a different tone quality on and off axis. Small-diaphragm mics
tend to have less off-axis coloration than large-diaphragm mics.
You can get either the condenser or dynamic type with any kind of
polar pattern (except bidirectional dynamic). Ribbon mics are either
bidirectional or hypercardioid. Some condenser mics come with
switchable patterns. Note that the shape of a mic doesn’t indicate its
polar pattern.
If a mic is end-addressed, you aim the end of the mic at the sound
source. If a mic is side-addressed, you aim the side of the mic at the
sound source. Figure 6.6 shows a typical side-addressed condenser
mic with switchable polar patterns.
Boundary mics that mount on a surface have a pattern that is halfomni (hemispherical), half-supercardioid, or half-cardioid (like
an apple sliced in half through its stem). The boundary mounting
makes the mic more directional so it picks up less room acoustics.
Microphones CHAPTER 6
DIAPHRAGM
GRILLE
ELECTRONICS
FIGURE 6.6
A typical multi-pattern mic that is side-addressed.
FREQUENCY RESPONSE
As with other audio components, a microphone’s frequency response
is the range of frequencies that it will reproduce at an equal level
(within a tolerance, such as ⫾3 dB).
The following is a list of sound sources and the microphone frequency response that is adequate to record the source with high
fidelity. A wider range response works, too.
■
■
■
■
■
■
Most instruments: 80 Hz to 15 kHz
Bass instruments: 40 Hz to 9 kHz
Brass and voice: 80 Hz to 12 kHz
Piano: 40 Hz to 12 kHz
Cymbals and some percussion: 300 Hz to 15 or 20 kHz
Orchestra or symphonic band: 40 Hz to 15 kHz
If possible, use a mic with a response that rolls off below the lowest fundamental frequency of the instrument you’re recording. For
example, the frequency of the low-E string on an acoustic guitar is
about 82 Hz. A mic used on the acoustic guitar should roll off below
that frequency to avoid picking up low-frequency noise such as rumble from trucks and air conditioning. Some mics have a built-in lowcut switch for this purpose. Or you can filter out the unneeded lows
at your mixer.
87
88
CHAPTER 6 Microphones
+10
dB
2 IN. FROM
SOUND SOURCE
0
2 FT. FROM
SOUND SOURCE
–10
100
500
1K
5K
10K 20K
FREQUENCY IN Hz
FIGURE 6.7
An example of the frequency response of a microphone with proximity effect and a presence peak
around 5 kHz.
A frequency-response curve is a graph of the mic’s output level in dB
at various frequencies. The output level at 1 kHz is placed at the 0 dB
line on the graph, and the levels at other frequencies are so many
decibels above or below that reference level.
The shape of the response curve suggests how the mic sounds at a
certain distance from the sound source. (If the distance isn’t specified, it’s probably 2 to 3 feet.) For example, a mic with a wide, flat
response reproduces the fundamental frequencies and harmonics
in the same proportion as the sound source. So a flat-response mic
tends to provide accurate, natural reproduction at that distance.
A rising high end or a “presence peak” around 5 to 10 kHz sounds
more crisp and articulate because it emphasizes the higher harmonics (Figure 6.7). Play audio clip 15 at www.elsevierdirect.com/
companions/9780240811444. Sometimes this type of response is
called tailored or contoured. It’s popular for guitar amps and drums
because it adds punch and emphasizes attack. Some microphones
have switches that alter the frequency response.
Most uni- and bidirectional mics boost the bass when used within a
few inches of a sound source. You’ve heard how the sound gets bassy
when a vocalist sings right into the mic. This low-frequency boost
related to close mic placement is called the proximity effect, and it’s
often plotted on the frequency-response graph. Omni mics have no
proximity effect; they sound tonally the same at any distance.
The warmth created by proximity effect adds a pleasing fullness to
drums. In most recording situations, though, the proximity effect
Microphones CHAPTER 6
lends an unnatural boomy or bassy sound to the instrument or
voice picked up by the mic. Some mics—multiple-D or variable-D
types—are designed to reduce it. These types have holes or slots in
the mic handle. Some mics have a bass-rolloff switch to compensate for the bass boost. Or you can roll off the excess bass with your
mixer’s equalizer until the sound is natural. By doing so, you also
reduce low-frequency leakage picked up by the microphone.
Note that mic placement can greatly affect the recorded tone quality. A flat-response mic doesn’t always guarantee a natural sound
because mic placement has such a strong influence. Tonal effects of
mic placement are covered in Chapter 7.
IMPEDANCE (Z)
This spec is the mic’s effective output resistance at 1 kHz. A mic
impedance between 150 and 600 ohms is low; 1000 to 4000 ohms
is medium; and above 25 kilohms is high.
Always use low-impedance mics. If you do, you can run long mic
cables without picking up hum or losing high frequencies. The
input impedance of a mixer mic input is about 1500 ohms. If it were
the same impedance as the mic, about 250 ohms, the mic would “load
down” when you plug it in. Loading down a mic makes it lose level,
distort, or sound thin. To prevent this, a mic input has an impedance much higher than that of the microphone. But it’s still called a
low-Z input.
More information on impedance is in Appendix C.
MAXIMUM SPL
To understand this spec, first we need to understand sound pressure
level (SPL). It is a measure of the intensity of a sound. The quietest sound we can hear, the threshold of hearing, is 0 dBSPL. Normal
conversation at 1 foot measures about 70 dBSPL; painfully loud
sound is above 120 dBSPL.
If the maximum SPL spec is 125 dBSPL, the mic starts to distort when
the instrument miked is putting out 125 dBSPL at the mic. A maximum SPL spec of 120 dB is good, 135 dB is very good, and 150 dB is
excellent.
89
90
CHAPTER 6 Microphones
Dynamic mics tend not to distort, even with very loud sounds. Some
condensers are just as good. Some have a pad you can switch in
to prevent distortion in the mic circuitry. Because a mic pad reduces
signal-to-noise ratio (S/N), use it only if the mic distorts.
SENSITIVITY
This spec tells how much output voltage a mic produces when driven
by a certain SPL. A high-sensitivity mic puts out a stronger signal
(higher voltage) than a low-sensitivity mic when both are exposed to
an equally loud sound.
A low-sensitivity mic needs more mixer gain than a high-sensitivity
mic. More gain usually results in more noise. When you record quiet
music at a distance (classical guitar, string quartet), use a mic of high
sensitivity to override mixer noise. When you record loud music
or mike close, sensitivity matters little because the mic signal level
is well above the mixer noise floor; that is, the S/N is high. Listed
below are typical sensitivity specs for three transducer types:
■
■
■
Condenser: 5.6 mV/Pa (high sensitivity)
Dynamic: 1.8 mV/Pa (medium sensitivity)
Ribbon or small dynamic: 1.1 mV/Pa (low sensitivity)
The louder the sound source, the higher the signal voltage the mic puts
out. A very loud instrument, such as a kick drum or guitar amp, can
cause a microphone to generate a signal strong enough to overload the
mic preamp in your mixer. That’s why most mixers have pads or inputgain controls—to prevent preamp overload from hot mic signals.
SELF-NOISE
Self-noise or equivalent noise level is the electrical noise or hiss a
mic produces. It’s the dBSPL of a sound source that would produce
the same output voltage that the noise does.
Usually the self-noise spec is A-weighted. That means the noise was
measured through a filter that makes the measurement correlate
more closely with the annoyance value. The filter rolls off low and
high frequencies to simulate the frequency response of the ear.
An A-weighted self-noise spec of 14 dBSPL or less is excellent (quiet);
21 dB is very good, 28 dB is good; and 35 dB is fair—not good
enough for quality recording.
Microphones CHAPTER 6
Because a dynamic mic has no active electronics to generate noise, it
has very low self-noise (hiss). So most spec sheets for dynamic mics
don’t specify self-noise.
S/N (SIGNAL-TO-NOISE RATIO)
This is the difference in decibels between the mic’s sensitivity and
its self-noise. The higher the SPL of the sound source at the mic,
the higher the S/N. Given an SPL of 94 dB, an S/N spec of 74 dB is
excellent; 64 dB is good. The higher the S/N ratio, the cleaner (more
noise-free) the signal, and the greater the “reach” of the microphone.
Reach is the clear pickup of quiet, distant sounds due to high S/N.
Reach isn’t specified in data sheets because any mic can pick up a
source at any distance if the source is loud enough. For example,
even a cheap mic can reach several miles if the sound source is a
thunderclap.
POLARITY
The polarity spec relates the polarity of the electrical output signal
to the acoustic input signal. The standard is “pin 2 hot.” That is, the
mic produces a positive voltage at pin 2 with respect to pin 3 when
the sound pressure pushes the diaphragm in (positive pressure).
Be sure that your mic cables don’t reverse polarity. On both ends of
each cable, the wiring should be pin 1 shield, pin 2 red, pin 3 white
or black. Or the wiring on both ends should be pin 1 shield, pin 2
white, pin 3 black. If some mic cables are correct polarity and some
are reversed, and you mix their mics to mono, the bass may cancel.
MICROPHONE TYPES
The following sections describe several types of recording mics.
Large-Diaphragm Condenser Microphone
This is a condenser microphone, usually side-addressed, with a diaphragm 1 inch or larger in diameter (Figure 6.6). It generally has
very good low-frequency response and low self-noise. Common
uses are studio vocals and acoustic instruments. Examples include:
AKG C12VR, C414, Perception 100 and 200, C2000B and C3000B;
Audio-Technica AT2020/3035/4040, Audix SCX25, Blue Blueberry,
91
92
CHAPTER 6 Microphones
CAD Equitek Series and M177, DPA 4041, Lawson L47MP MKII and
L251, Manley Gold Reference, Neumann U87, U47 and TLM 103;
Soundelux Elux 251, Sterling Audio ST51, ADK various models, SE
Electronics various models, Shure KSM Series, MXL V67G, V69, 900,
2001, 2003 and 2006; Rode NT1A, Studio Projects B and C Series,
Samson CL7 and C01, Nady SCM 950 and 100, Violet Flamingo,
M-Audio Solaris, Sputnik, Luna and Nova; and Behringer B1 and B2.
Small-Diaphragm Condenser Microphone
This is a stick-shaped or “pencil” condenser microphone, usually cardioid and end-addressed, with a diaphragm under 1 inch in diameter
(Figure 6.4). It generally has very good transient response and detail,
making it a fine choice for close miking acoustic instruments—
especially cymbals, acoustic guitar, and piano. Examples include:
AKG C 451 B; Audio-Technica AT 3031 and AT 4051a; Audix SCX1,
ADX50, and ADX51; Berliner CM-33, CAD Equitek e60; M-Audio
Pulsar; Samson C02; Crown CM-700; DPA 4006; Mojave MA-100,
Neumann KM 184; Sennheiser e614 and MKH50; Shure KSM109/SL,
KSM137/Sl and SM81; MXL 600 and 603S; Behringer B5 with cardioid capsule; and Studio Projects C4.
Dynamic Instrument Microphone
This is a stick-shaped dynamic microphone, end-addressed (Figure 6.4).
Although it may have a flat response, it generally has a presence peak
and some low-frequency rolloff to prevent boominess when used up
close. It’s often used on drums and guitar amps. Examples include:
Shure SM57, AKG D112 (kick drum), Audio-Technica AT AE2500
(kick), Electro-Voice N/D868 (kick), Heil PR40 (kick), Audix D1
through D6 and I-5, and Sennheiser MD421, e604 and e602 (kick).
Live-Vocal Microphone
This unidirectional mic is shaped like an ice-cream cone because
of its large grille used to reduce breath pops. It can be a condenser,
dynamic, or ribbon type, and it usually has a presence peak and
some low-frequency rolloff. Examples include: Shure SM58, Beta 58,
SM85, SM86, SM87A and KSM9; AKG D3800, C 535 EB, C5, ELLE C,
and D 5/D 5 S; Audix OM5, Beyerdynamic M88 TG, Crown CM200A, EV N-Dym Series, and Neumann KMS104 and KMS105.
Microphones CHAPTER 6
Ribbon Microphone
This mic can be side- or end-addressed. It generally is used wherever
you want a warm, smooth tone quality (sometimes with reduced
highs). Examples include: models by Beyerdynamic, Coles, Royer,
Cascade, Blue, and AEA.
Boundary Microphone
Boundary mics are designed to be used on surfaces. Tape them to the
underside of a piano lid, or tape them to the wall for pickup of room
ambience. They can be used on hard baffles between instruments, or
on panels to make the mics directional. A boundary mic uses a mini
condenser mic capsule mounted very near a sound-reflecting plate
or boundary (Figure 6.8). Because of this construction, the mic picks
up direct sound and reflected sound at the same time, in-phase at
all frequencies. So you get a smooth response free of phase cancellations. A conventional mic near a surface sounds colored; a boundary
mic on a surface sounds natural. Examples include: AKG C 562 BL,
Audio-Technica AT 841a, Beyerdynamic MPC 22, Crown PZM-30D
and PZM-6D, and Shure Beta 91.
MIC CAPSULE
CAPSULE
HOLDER
BOUNDARY PLATE
CAPSULE
HOLDER
SIDE VIEW
PLATE
FIGURE 6.8
Typical PZM construction.
MIC CAPSULE
93
94
CHAPTER 6 Microphones
1c
FIGURE 6.9
A mini mic is the size of a penny.
Other benefits are a wide, smooth frequency response free of phase
cancellations, excellent clarity and reach, and the same tone quality
anywhere around the mic. The polar pattern is half-omni or hemispherical. Some boundary mics have a half-cardioid or half-supercardioid polar pattern. They work great on a conference table, or near
the front edge of a stage floor to pick up drama or musicals.
Miniature Microphone
Mini condenser mics can be attached to drum rims, flutes, horns,
acoustic guitars, fiddles, and so on. Their tone quality is about as
good as larger studio microphones and the price is relatively low.
With these tiny units you can mike a band in concert without cluttering the stage with boom stands (Figure 6.9), or you can mike a whole
drum set with two or three of these. Although you lose individual control of each drum in the mix, the cost is low and the sound is quite
good with some bass and treble boost. Compared to large mics, mini
mics tend to have more noise (hiss) in distant-miking applications.
A lavalier mic is a mini mic worn on the chest to pick up speech from
a newscaster or a wandering lecturer. Examples include: AKG Micro
Mic Series, Shure Beta 98S, Audix M1245 and Micro-D, Countryman
Isomax B6, Crown GLM-100, DPA 4060, and Sennheiser e608.
Microphones CHAPTER 6
FIGURE 6.10
Audio-Technica AT825, an example of a stereo microphone.
Stereo Microphone
A stereo microphone combines two directional mic capsules in a
single housing for convenient stereo recording (Figure 6.10). Simply
place the mic a suitable distance and height from the sound source,
and you’ll get a stereo recording with little fuss. Examples include:
AKG C426 B Comb, Audio-Technica AT825 and AT822, Crown
SASS-P MKII, Neumann SM69, Shure VP88, Nady RSM-2, AEA R88,
and Royer SF-12.
Because there is no spacing between the mic capsules, there also
is no delay or phase shift between their signals. Coincident stereo
microphones are mono-compatible—the frequency response is the
same in mono and stereo—because there are no phase cancellations
if the two channels are combined.
Digital Microphone
This condenser microphone has a built-in A/D converter. It is usually side-addressed, has a large diaphragm, has a flat response, and
very low self-noise. Its output is a digital signal, which is immune
to picking up hum. Examples include: Beyerdynamic MCD 100 and
Neumann Solution-D.
95
96
CHAPTER 6 Microphones
Headworn Microphone
This microphone is used for a live performance that might be
recorded. It’s a small condenser mic worn on the head, either omnior unidirectional. The headworn mic allows the performer freedom
of movement on stage. Some models provide excellent gain before
feedback and isolation. Examples include: AKG C 420, AudioTechnica ATM73, Countryman Isomax E6, DPA 4065, and Crown
CM-311A.
MICROPHONE SELECTION
Table 6.1 is a guide to choosing a mic based on your requirements.
Suppose you want to record a grand piano playing with several other
instruments. You need the microphone to reduce leakage. Table 6.1
recommends a unidirectional mic or an omni mic up close. For this
particular piano, you also want a natural sound, for which the table
suggests a mic with a flat response. You want a detailed sound, so a
condenser mic is the choice. A microphone with all these characteristics is a flat-response, unidirectional condenser mic. If you’re miking
close to a surface (the piano lid), a boundary mic is recommended.
Now suppose you’re recording an acoustic guitar on stage, and the
guitarist roams around. This is a moving sound source, for which the
table recommends a mini mic attached to the guitar. Feedback and
leakage might not be a problem because you’re miking close, so you
can use an omni mic. Thus, an omni condenser mic is a good choice
for this application.
For a home studio, a suggested first choice is a cardioid condenser
mic with a flat frequency response. This type of mic is especially
good for studio vocals, cymbals, percussion, and acoustic instruments. Remember that the mic needs a power supply to operate,
such as a battery or phantom power supply.
Your second choice of microphone for a home studio is a cardioid dynamic microphone with a presence peak in the frequency
response. This type is good for drums and guitar amps. I recommend
cardioid over omni for a home studio. The cardioid pattern rejects
the leakage, background noise, and room reverb often found in
home studios. An omni mic, however, can do that, too, if you mike
Microphones CHAPTER 6
Table 6.1
Mic Application Guide
Requirement
Characteristic
Natural, smooth tone quality
Flat frequency response
Bright, present tone quality
Presence peak (emphasis around 5 kHz)
Extended lows
Omni condenser or dynamic with good
low-frequency response
Extended highs (detailed sound)
Condenser
Reduced “edge” or detail
Dynamic or ribbon
Boosted bass up close
Directional mic
Flat bass response up close
Omni mic, or directional mic with
sound holes in the handle
Reduced pickup of leakage,
feedback, and room acoustics
Directional mics
Enhanced pickup of room
acoustics
Omni mics
Miking close to a surface, even
coverage of moving sources or
large sources, inconspicuous mic
Boundary mic
Coincident or near-coincident
stereo (see Chapter 18)
Stereo mic
Extra ruggedness
Dynamic mic
Reduced handling noise
Omni mic, or unidirectional with shock
mount
Reduced breath popping
Omni mic, or unidirectional with pop
filter
Distortion-free pickup of very
loud sounds
Condenser with high maximum SPL
spec, or dynamic
Low self-noise, high sensitivity,
noise-free pickup of quiet sounds
Large-diaphragm condenser mic
close enough. Also, omni mics tend to provide a more natural sound
at lower cost, and they have no proximity effect.
MIC ACCESSORIES
There are many devices used with microphones to route their signals or to make them more useful. These include pop filters, stands
97
98
CHAPTER 6 Microphones
and booms, shock mounts, cables and connectors, stage boxes and
snakes, and splitters.
Pop Filter
A much needed accessory for a vocalist’s microphone is a pop filter
or windscreen. It usually is a foam “sock” that you put over the mic.
Some microphones have pop filters or ball-shaped grilles built in.
Why is it needed? When a vocalist sings a word starting with “p,”
“b,” or “t” sounds, a turbulent puff of air is forced from the mouth.
A microphone placed close to the mouth is hit by this air puff,
resulting in a thump or little explosion called a pop. The windscreen
reduces this problem.
The best type of pop filter is a nylon screen in a hoop, or a perforated-metal disk, placed a few inches from the mic.
You can also reduce pop by placing the mic above or to the side of
the mouth, or by using an omni mic. Audio clip 16 at www.elsevierdirect
.com/companions/9780240811444 demonstrates how a pop filter or mic
placement can prevent breath pops.
Stands and Booms
Stands and booms hold the microphones and let you position them
as desired. A mic stand has a heavy metal base that supports a vertical pipe. At the top of the pipe is a rotating clutch that lets you adjust
the height of a smaller telescoping pipe inside the large one. The top
of the small pipe has a standard 5/8-inch 27 thread, which screws
into a mic stand adapter.
A boom is a long horizontal pipe that attaches to the vertical pipe.
The angle and length of the boom are adjustable. The end of the
boom is threaded to accept a mic stand adapter, and the opposite
end is weighted to balance the weight of the microphone.
Shock Mount
A shock mount holds a mic in a resilient suspension to isolate the
mic from mechanical vibrations, such as floor thumps and mic-stand
bumps. Many mics have an internal shock mount which isolates the
mic capsule from its housing; this reduces handling noise as well as
stand thumps.
Microphones CHAPTER 6
Cables and Connectors
Mic cables carry the electrical signal from the mic to the mixing console, mic preamp, or recorder. With low-impedance mics, you can
use hundreds of feet of cable without hum pickup or high-frequency
loss. Some mics have a permanently attached cable for convenience
and low cost; others have a connector in the handle to accept a separate mic cable. The second method is preferred for serious recording
because if the cable breaks, you have to repair or replace only the
cable, not the whole microphone.
Mic cables are made of one or two insulated conductors surrounded
by a fine-wire mesh shield to keep out electrostatic hum. If you hear
a loud buzz when you plug in a microphone, check that the shield is
securely soldered to XLR pin 1 on both ends of the cable.
After acquiring a microphone, you may need to wire its 2-conductor
shielded cable to a 3-pin XLR audio connector. Here are the solder
connections:
■
■
■
Pin 1: Shield
Pin 2: “Hot” or “in-polarity” lead (usually red or white)
Pin 3: “Cold” or “out-of-polarity” lead (usually black)
If the mic output is a 3-pin XLR, but your recorder or mixer mic
input is an unbalanced phone jack, a different wiring is needed:
■
■
Phone-plug tip (the short terminal): Hot lead
Phone-plug sleeve (the long terminal): Shield and cold lead
Wind your mic cables onto a large spool, which can be found in the
electrical section of hardware stores. Plug the cables together as you
wind them.
Snake
It’s messy and time-consuming to run mic cables from several mics
all the way to a mixer. Instead, you can plug all your mics into a
stage box with several connectors (Figure 6.11). The snake—a thick
multi-conductor cable—carries the signals to the mixer. At the mixer
end, the cable divides into several mic connectors that plug into the
mixer.
99
100
CHAPTER 6 Microphones
STAGE BOX
CONNECTORS TO
MIC CABLES
SNAKE
CONNECTORS TO MIXER
FIGURE 6.11
A stage box and snake.
Splitter
When you record a band in concert, you might want to feed each
mic’s signal to your recording mixer and to the band’s PA and monitor mixers. A mic splitter does the job. For each microphone channel,
it has one XLR input for a microphone, a “direct” XLR output wired
to the input, and one or more transformer-isolated XLR outputs with
a ground-lift switch. The mixer that provides phantom power must
be connected to the direct XLR output. There’s more on splitters in
Chapter 17 under the heading Splitting the Mic Signals.
SUMMARY
We talked about some mic types, specs, and accessories. You should
have a better idea about what kind of microphone to choose for your
own applications.
Mic manufacturers are happy to send you free catalogs and application notes, which are also available on mic company Web sites. Mic
dealers also may have this literature.
Remember, you can use any microphone on any instrument if it
sounds good to you. Just try it and see if you like it. To make highquality recordings, though, you need good mics with a smooth,
wide-range frequency response, low noise, and low distortion.
CHAPTER 7
Microphone Technique
Basics
Suppose you’re going to mike a singer, a sax, or a guitar. Which mic
should you choose? Where should you place it?
Your mic technique has a powerful effect on the sound of your
recordings. In this chapter we’ll look at some general principles of
miking that apply to all situations. Chapter 8 covers common mic
techniques for specific instruments.
WHICH MIC SHOULD I USE?
Is there a “right” mic to use on a piano, a kick drum, or a guitar
amp? No. Every microphone sounds different, and you choose the
one that gives you the sound you want. Still, it helps to know about
two main characteristics of mics that affect the sound: frequency
response and polar pattern.
Most condenser mics have an extended high-frequency response—
they reproduce sounds up to 15 or 20 kHz. This makes them great
for cymbals or other instruments that need a detailed sound, such
as acoustic guitar, strings, piano, and voice. Dynamic moving-coil
microphones have a response good enough for drums, guitar amps,
horns, and woodwinds. Loud drums and guitar amps sound dull
if recorded with a flat-response mic; a mic with a presence peak
(a boost around 5 kHz) gives more edge or punch.
Suppose you’re choosing a microphone for a particular instrument.
In general, the frequency response of the mic should cover at least
the frequencies produced by that instrument. For example, an acoustic guitar produces fundamental frequencies from 82 Hz to about
101
102
CHAPTER 7 Microphone Technique Basics
1 kHz, and produces harmonics from about 1 to 15 kHz. So a mic
used on an acoustic guitar should have a frequency response of at
least 82 Hz to 15 kHz if you want to record the guitar accurately.
Table 7.1 shows the frequency ranges of various instruments.
The polar pattern of a mic affects how much leakage and ambience
it picks up. Leakage is unwanted sound from instruments other than
the one at which the mic is aimed. Ambience is the acoustics of the
Table 7.1
Frequency Ranges of Various Musical Instruments
Instrument
Fundamentals (Hz)
Harmonics (kHz)
Flute
261–2349
3–8
Oboe
261–1568
2–12
Clarinet
165–1568
2–10
Bassoon
62–587
1–7
Trumpet
165–988
1–7.5
French horn
87–880
1–6
Trombone
73–587
1–7.5
Tuba
49–587
1–4
Snare drum
100–200
1–20
Kick drum
30–147
1–6
Cymbals
300–587
1–15
Violin
196–3136
4–15
Viola
131–1175
2–8.5
Cello
65–698
1–6.5
Acoustic bass
41–294
1–5
Electric bass
41–300
1–7
Acoustic guitar
82–988
1–15
Electric guitar
82–1319
1–3.5 (through
amp)
Electric guitar
82–1319
1–15 (direct)
Piano
28–4196
5–8
Bass (voice)
87–392
1–12
Tenor (voice)
131–494
1–12
Alto (voice)
175–698
2–12
Soprano (voice)
247–1175
2–12
Microphone Technique Basics CHAPTER 7
recording room—its early reflections and reverb. The more leakage
and ambience you pick up, the more distant the instrument sounds.
An omni mic picks up more ambience and leakage than a directional mic when both are the same distance from an instrument. So
an omni tends to sound more distant. To compensate, you have to
mike closer with an omni.
HOW MANY MICS?
The number of mics you need varies with what you’re recording. If you
want to record an overall acoustic blend of the instruments and room
ambience, use just two microphones or a stereo mic (Figure 7.1). This
method works great on an orchestra, symphonic band, choir, string
quartet, pipe organ, small folk group, or a piano/voice recital. Stereo
miking is covered in detail later in this chapter under the heading Stereo
Mic Techniques.
To record a pop-music group, you mike each instrument or instrumental section. Then you adjust the mixer volume control for each
mic to control the balance between instruments (Figure 7.2).
To get the clearest sound, don’t use two mics when one will do the
job. Sometimes you can pick up two or more sound sources with one
mic (Figure 7.3). You could mike a brass section of four players with
FIGURE 7.1
Overall miking of a musical ensemble with two distant microphones.
103
104
CHAPTER 7 Microphone Technique Basics
FIGURE 7.2
Individual miking with multiple close mics and a mixer.
FIGURE 7.3
Multiple miking with several sound sources on each microphone.
one mic on four players, or with two mics on every two players. Or
mike a choir in a studio in four groups: put one mic on the basses,
one on the sopranos, and so on.
Picking up more than one instrument with one mic has a problem:
during mixdown, you can’t adjust the balance among instruments
recorded on the same track. You have to balance the instruments
before recording them. Monitor the mic, and listen to see if any
instrument is too quiet. If so, move it closer to the mic.
Microphone Technique Basics CHAPTER 7
HOW CLOSE SHOULD I PLACE THE MIC?
Once you’ve chosen a mic for an instrument, how close should the
mic be? Mike a few inches away to get a tight, present sound; mike
farther away for a distant, spacious sound. (Try it to hear the effect.)
Play audio clip 17 at www.elsevierdirect.com/companions/9780240811444.
The farther a mic is from the instrument, the more ambience, leakage, and background noise it picks up. So mike close to reject these
unwanted sounds. Mike farther away to add a live, loose, airy feel to
overdubs of drums, lead-guitar solos, horns, etc.
Close miking sounds close; distant miking sounds distant. Here’s
why. If you put a mic close to an instrument, the sound at the mic is
loud. So you need to turn up the mic gain on your mixer only a little
to get a full recording level. And because the gain is low, you pick up
very little reverb, leakage, and background noise (Figure 7.4A).
If you put a mic far from an instrument, the sound at the mic is
quiet. You’ll need to turn up the mic gain a lot to get a full recording
level. And because the gain is high, you pick up a lot of reverb, leakage, and background noise (Figure 7.4B).
A
REVERB
REFLECTIONS
MIC
NOISE
B
REVERB
REFLECTIONS
MIC
NOISE
FIGURE 7.4
(A) A close microphone picks up mainly direct sound, which results in a close sound quality. (B) A distant
microphone picks up mainly reflected sound, which results in a distant sound quality.
105
106
CHAPTER 7 Microphone Technique Basics
If the mic is very far away—maybe 10 feet—it’s called an ambience
mic or room mic. It picks up mostly room reverb. A popular mic for
ambience is a boundary microphone taped to the wall. You mix it
with the usual close mics to add a sense of space. Use two for stereo.
When you record a live concert, you might want to place ambience
mics over the audience, aiming at them from the front of the hall, to
pick up the crowd reaction and the hall acoustics.
Classical music is always recorded at a distance (about 4 to 20 feet
away) so that the mics will pick up reverb from the concert hall. It’s
a desirable part of the sound.
Leakage (Bleed or Spill)
Suppose you’re close-miking a drum set and a piano at the same
time (Figure 7.5). When you listen to the drum mics alone, you
hear a close, clear sound. But when you mix in the piano mic, that
nice, tight drum sound degrades into a distant, muddy sound. That’s
because the drum sound leaked into the piano mic, which picked
up a distant drum sound from across the room. Audio clip 11 at www
.elsevierdirect.com/companions/9780240811444 is an example of leakage.
There are many ways to reduce leakage:
■
■
Mike each instrument closely. That way the sound level at each
mic is high. Then you can turn down the mixer gain of each
mic, which reduces leakage at the same time.
Overdub each instrument one at a time.
DRUMS
PIANO
DRUM LEAKAGE
PIANO MIC
DIRECT
DRUM
SOUND
DRUM MIC(S)
FIGURE 7.5
Example of leakage. The piano mic picks up leakage from the drums, which changes the close drum
sound to distant.
Microphone Technique Basics CHAPTER 7
■
■
■
■
■
■
Record direct. Record an acoustic guitar off its pickup during
tracking, then overdub the guitar with a mic. Record an electric
guitar off its pickup during tracking, then play the guitar signal
through a guitar-amp modeling plug-in during mixdown. Or
record the electric guitar through a Line 6 Pod, which is a guitaramp emulator.
Filter out frequencies above and below the range of each
instrument.
Use directional mics (cardioid, etc.) instead of omni mics.
Record in a large, fairly dead studio. In such a room, leakage
reflected from the walls is weak.
Put portable walls (goboes) between instruments.
Use noise gates on drum tracks (see Chapter 10 under the heading Noise Gate.)
Don’t Mike Too Close
Miking too close can color the recorded tone quality of an instrument. If you mike very close, you might hear a bassy or honky tone
instead of a natural sound. Why? Most musical instruments are
designed to sound best at a distance, at least 1-1/2 feet away. The
sound of an instrument needs some space to develop. A mic placed
a foot or two away tends to pick up a well-balanced, natural sound.
That is, it picks up a blend of all the parts of the instrument that contribute to its character or timbre.
Think of a musical instrument as a loudspeaker with a woofer, midrange, and tweeter. If you place a mic a few feet away, it will pick
up the sound of the loudspeaker accurately. But if you place the mic
close to the woofer, the sound will be bassy. Similarly, if you mike
close to an instrument, you emphasize the part of the instrument
that the microphone is near. The tone quality picked up very close
may not reflect the tone quality of the entire instrument.
Suppose you place a mic next to the sound hole of an acoustic guitar,
which resonates around 80 to 100 Hz. A microphone placed there
hears this bassy resonance, giving a boomy recorded timbre that
doesn’t exist at a greater miking distance. To make the guitar sound
more natural when miked close to the sound hole, you need to roll
off the excess bass on your mixer, or use a mic with a bass rolloff in
its frequency response.
107
108
CHAPTER 7 Microphone Technique Basics
The sax projects highs from the bell, but projects mids and lows
from the tone holes. So if you mike close to the bell, you miss the
warmth and body from the tone holes. All that’s left at the bell is a
harsh tone quality. You might like that sound, but if not, move the
mic out and up to pick up the entire instrument. If leakage forces
you to mike close, change the mic or use equalization (EQ).
Usually, you get a natural sound if you put the mic as far from the
source as the source is big. That way, the mic picks up all the soundradiating parts of the instrument about equally. For example, if the
body of an acoustic guitar is 18 inches long, place the mic 18 inches
away to get a natural tonal balance. If this sounds too distant or hollow, move in a little closer.
WHERE SHOULD I PLACE THE MIC?
Suppose you have a mic placed a certain distance from an instrument. If you move the mic left, right, up, or down, you change the
recorded tone quality. In one spot, the instrument might sound bassy;
in another spot, it might sound natural, and so on. So, to find a good
mic position, simply place the mic in different locations—and monitor the results—until you find one that sounds good to you.
Here’s another way to do the same thing. Close one ear with your
finger, listen to the instrument with the other ear, and move around
until you find a spot that sounds good. Put the mic there. Then
make a recording and see if it sounds the same as what you heard
live. Don’t try this with kick drums or screaming guitar amps! You
could also move a mic around while monitoring its signal with good
headphones.
Why does moving the mic change the tone quality? A musical instrument radiates a different tone quality in each direction. Also, each
part of the instrument produces a different tone quality. For example,
Figure 7.6 shows the tonal balances picked up at various spots
near a guitar. Audio clip 18 at www.elsevierdirect.com/companions/
9780240811444 illustrates the effect of mic placement on guitar tonal balance. Audio clip 19 demonstrates close and distant stereo miking of the
acoustic guitar.
Other instruments work the same way. A trumpet radiates strong
highs directly out of the bell, but doesn’t project them to the sides.
Microphone Technique Basics CHAPTER 7
NATURAL
NATURAL
BASSY
MELLOW
FIGURE 7.6
Microphone placement affects the recorded tonal balance.
So a trumpet sounds bright when miked on-axis to the bell and
sounds more natural or mellow when miked off to one side. A grand
piano miked one foot over the middle strings sounds fairly natural,
under the soundboard sounds bassy and dull, and in a sound hole it
sounds mid-rangey.
It pays to experiment with all sorts of mic positions until you find a
sound you like. There is no one right way to place the mics because
you place them to get the tonal balance you want.
ON-SURFACE TECHNIQUES
Sometimes you’re forced to place a mic near a hard reflecting surface.
Examples:
■
■
■
Recording drama or opera with the mics near the stage floor
Recording an instrument that has hard surfaces around it
Recording a piano with the mic close to the lid
In these cases, you’ll often pick up an unnatural, filtered tone quality. Here’s why. Sound travels to the microphone via two paths:
directly from the sound source, and reflected off the nearby surface
(Figure 7.7). Because of its longer travel path, the reflected sound is
delayed compared to the direct sound. The direct and delayed sound
109
110
CHAPTER 7 Microphone Technique Basics
SOUND
SOURCE
DIRECT
MICROPHONE
REFLECTED
SURFACE
dB
FREQUENCY
FIGURE 7.7
A mic placed near a surface picks up direct sound and a delayed reflection, which gives a comb-filter
frequency response.
waves combine at the mic, which causes phase cancellations of various frequencies. The series of peaks and dips in the response is called
a comb-filter effect, and it sounds like mild flanging.
Boundary mics solve the problem. In a boundary mic, the diaphragm
is very close to the reflecting surface so that there is no delay in the
reflected sound. Direct and reflected sounds add in-phase over the
audible range of frequencies, resulting in a flat response (Figure 7.8).
Play audio clip 20 at www.elsevierdirect.com/companions/9780240811444.
You might tape an omni boundary mic to the underside of a piano
lid, to a hard-surfaced panel, or to a wall for ambience pickup.
A unidirectional boundary mic works great on a stage floor to pick
up drama. A group of these mics will clearly pick up people at a conference table.
THE 3:1 RULE
Suppose you’re recording a singer/guitarist. There’s a mic on the
singer and a mic on the acoustic guitar. When you monitor the mix,
something’s wrong: the singer’s voice sounds hollow or filtered.
You’re hearing the effect of phase interference.
Microphone Technique Basics CHAPTER 7
CLOSE-UP
VIEW
SOUND
SOURCE
MIC CAPSULE
REFLECTED
SOUND
BOUNDARY MICROPHONE
dB
FREQUENCY
FIGURE 7.8
A boundary mic on the surface picks up direct and reflected sounds in phase.
In general, if two mics pick up the same sound source at different distances, and their signals are mixed to the same channel, this
might cause phase cancellations. These are peaks and dips in the frequency response—a comb filter—caused by some frequencies combining out of phase. The result is a colored, filtered tone quality that
sounds like mild flanging.
To prevent this problem, follow the 3:1 rule: Space the mics at least
three times the mic-to-source distance (as in Figure 7.9). For example, if two mics are 12 inches apart, they should be less than 4 inches
from their sound sources to prevent phase cancellations.
Play audio clip 21 at www.elsevierdirect.com/companions/9780240811444.
The mics can be closer together than 3:1 if you use two cardioid mics
aiming in opposite directions. The goal is to get at least 9 dB of separation between recorded tracks.
What if you pick up an instrument with two mics that are panned
left and right? You don’t get phase interference. Instead you get stereo imaging.
111
112
CHAPTER 7 Microphone Technique Basics
3D OR GREATER
D
D
FIGURE 7.9
The 3:1 rule of microphone placement avoids phase interference between microphone signals.
OFF-AXIS COLORATION
Some mics have off-axis coloration—a dull or colored effect on
sound sources that aren’t directly in front of the mic. Try to aim the
mic at sound sources that put out high frequencies, such as cymbals. When you pick up a large source such as an orchestra, use a mic
that has the same response over a wide angle. Such a mic has similar
polar patterns at middle and high frequencies. Most large-diaphragm
mics have more off-axis coloration than smaller mics (under 1 inch).
STEREO MIC TECHNIQUES
Stereo mic techniques capture the sound of a musical group as a
whole, using only two or three microphones. When you play back
a stereo recording, you hear phantom images of the instruments in
various spots between the speakers. These image locations—left to
right, front to back—correspond to the instrument locations during
the recording session.
Stereo miking is the preferred way to record classical-music ensembles and soloists. In the studio, you can stereo-mike a piano, drum
set cymbals, vibraphone, harmony singers, or other large sound
sources.
Goals of Stereo Miking
One goal is accurate localization; that is, instruments in the center of the group are reproduced midway between the two speakers.
Instruments at the sides of the group are heard from the left or right
Microphone Technique Basics CHAPTER 7
C
(A) Orchestra instrument
locations (top view)
RC
LC
R
L
(B) Images localized accurately
between speakers (the listener's
perception)
RIGHT
SPEAKER
LEFT
SPEAKER
L
LC
RC
R
L LC C RC R
(C) Narrow stage effect
(D) Exaggerated separation
effect
C
L LC
C
RC R
FIGURE 7.10
Stereo localization effects. (A) Orchestra instrument locations (top view). (B) Images localized accurately
between speakers (the listener’s perception). (C) Narrow stage effect. (D) Exaggerated separation effect.
speaker. Instruments halfway to one side are heard halfway to one
side, and so on.
Figure 7.10 shows three stereo localization effects. Figure 7.10A shows
some instrument positions in an orchestra: left, left-center, center,
right-center, right. In Figure 7.10B, the reproduced images of these
instruments are accurately localized between the speakers. The stereo
spread, or stage width, extends from speaker to speaker. (You might
want to record a string quartet with a narrower spread.)
If you space or angle the mics too close together, you get a narrow
stage effect (Figure 7.10C). If you space or angle the mics too far
apart, you hear exaggerated separation (Figure 7.10D); that is, instruments halfway to one side are heard near the left or right speaker.
To judge stereo effects, you have to sit exactly between your monitor
speakers (the same distance from each). Sit as far from the speakers as the spacing between them. Then the speakers appear to be 60
degrees apart. This is about the same angle an orchestra fills when
viewed from a typical ideal seat in the audience (say, tenth row center). If you sit off-center, the images shift toward the side on which
you’re sitting and are less sharp.
113
114
CHAPTER 7 Microphone Technique Basics
Types of Stereo Mic Techniques
To make a stereo recording, you use one of these basic techniques:
■
■
■
■
Coincident pair (XY or MS)
Spaced pair (AB)
Near-coincident pair (ORTF, etc.)
Baffled pair (sphere, OSS, SASS, PZM wedge, etc.)
Let’s look at each technique.
COINCIDENT PAIR
With this method, you mount two directional mics with grilles touching, diaphragms one above the other, and angled apart (Figure 7.11).
For example, mount two cardioid mics with one grille above the other,
and angle them 120 degrees apart. You can use other patterns too:
supercardioid, hypercardioid, or bidirectional. The wider the angle
between mics, the wider the stereo spread. If the angle is too wide,
center images will be weak (there will be a “hole in the middle”).
How does this technique make images you can localize? A directional mic is most sensitive to sounds in front of the mic (on-axis)
and progressively less sensitive to sounds arriving off-axis. That is, a
directional mic puts out a high-level signal from the sound source
it’s aimed at, and produces lower-level signals from other sound
sources.
FIGURE 7.11
Coincident-pair technique.
Microphone Technique Basics CHAPTER 7
The coincident pair uses two directional mics that are angled symmetrically from the center line (Figure 7.11). Instruments in the
center of the group make the same signal from each mic. During
playback, you hear a phantom image of the center instruments midway between your speakers. That’s because identical signals in each
channel produce an image in the center.
If an instrument is off-center to the right, it is more on-axis to the
right-aiming mic than to the left-aiming mic. So the right mic will
produce a higher level signal than the left mic. During playback of
this recording, the right speaker will play at a higher level than the
left speaker. This reproduces the image off-center to the right—where
the instrument was during recording.
The coincident pair codes instrument positions into level differences
between channels. During playback, the brain decodes these level
differences back into corresponding image locations. A pan pot in
a mixing console works on the same principle. If one channel is 15
to 20 dB louder than the other, the image shifts all the way to the
louder speaker.
Suppose we want the right side of the orchestra to be reproduced at
the right speaker. That means the far-right musicians must produce
a signal level 20 dB higher from the right mic than from the left mic.
This happens when the mics are angled far enough apart. The correct
angle depends on the polar pattern.
Instruments partway off center produce interchannel level differences
less than 20 dB, so you hear them partway off center.
Listening tests have shown that coincident cardioid mics tend to
reproduce the musical group with a narrow stereo spread; that is, the
group doesn’t spread all the way between speakers.
A coincident-pair method with excellent localization is the Blumlein
array. It uses two bidirectional mics angled 90 degrees apart and facing the left and right sides of the group.
A special form of the coincident-pair technique is mid-side or MS
(Figure 7.12). In this method, a cardioid or omni mic faces the middle of the orchestra. A matrix circuit sums and differences the cardioid mic with a bidirectional mic aiming to the sides. This produces
left- and right-channel signals. You can remotely control the stereo
115
116
CHAPTER 7 Microphone Technique Basics
TO MUSICAL
ENSEMBLE
MID
MIC
–
+
SIDE MIC
RIGHT CHANNEL = MID + SIDE
LEFT CHANNEL = MID – SIDE
FIGURE 7.12
Mid-side (MS) technique.
spread by changing the ratio of the mid signal to the side signal. This
remote control is useful at live concerts, where you can’t physically
adjust the mics during the concert. MS localization can be accurate.
Chapter 18, under the heading Stereo-Spread Control, tells how to
replace the matrix with a computer DAW.
To make coincident recordings sound more spacious, boost the bass
4 dB (⫹2 dB at 600 Hz) in the L–R or side signal.
A recording made with coincident mics is mono-compatible; that
is, the frequency response is the same in mono or stereo. Because
the mics occupy almost the same point in space, there is no time
or phase difference between their signals. And when you combine
them to mono, there are no phase cancellations to degrade the frequency response. If you expect that your recordings will be heard
in mono (say, on TV), then you’ll probably want to use coincident
methods.
SPACED PAIR
Here, you mount two identical mics several feet apart and aim them
straight ahead (Figure 7.13). The mics can have any polar pattern,
but omni is most popular for this method. The greater the spacing
between mics, the greater the stereo spread.
How does this method work? Instruments in the center of the group
make the same signal from each mic. When you play back this
Microphone Technique Basics CHAPTER 7
MUSICAL
ENSEMBLE
SPACING
LEFT
CHANNEL
OUTPUT
RIGHT
CHANNEL
OUTPUT
FIGURE 7.13
Spaced-pair technique.
recording, you hear a phantom image of the center instruments midway between your speakers. If an instrument is off-center, it is closer
to one mic than the other, so its sound reaches the closer microphone before it reaches the other one. Both mics make about the
same signal, except that one mic signal is delayed compared with the
other.
If you send a signal to two speakers with one channel delayed, the
sound image shifts off center. With a spaced-pair recording, offcenter instruments produce a delay in one mic channel, so they are
reproduced off center.
The spaced pair codes instrument positions into time differences
between channels. During playback, the brain decodes these time
differences back into corresponding image locations.
A delay of 1.2 msec is enough to shift an image all the way to one
speaker. You can use this fact when you set up the mics. Suppose you
want to hear the right side of the orchestra from the right speaker.
The sound from the right-side musicians must reach the right mic
about 1.2 msec before it reaches the left mic. To make this happen,
space the mics about 2 to 3 feet apart. This spacing makes the correct
delay to place right-side instruments at the right speaker. Instruments
partway off center make interchannel delays less than 1.2 msec, so
they are reproduced partway off center.
If the spacing between mics is, say, 12 feet, then instruments that are
slightly off center produce delays between channels that are greater
than 1.2 msec. This places their images at the left or right speaker. I call
this “exaggerated separation” or a “ping-pong” effect (Figure 7.10D).
117
118
CHAPTER 7 Microphone Technique Basics
On the other hand, if the mics are too close together, the delays produced will be too small to provide much stereo spread. Also, the
mics will tend to emphasize instruments in the center because the
mics are closest to them.
To record a good musical balance of an orchestra, you might need to
space the mics about 10 or 12 feet apart. But then you get too much
separation. You could place a third mic midway between the outer
pair and mix its output to both channels. That way, you pick up a
good balance, and you hear an accurate stereo spread.
The spaced-pair method tends to make off-center images unfocused
or hard to localize. Why? Spaced-pair recordings have time differences between channels. Stereo images produced solely by time differences are unfocused. You still hear the center instruments clearly
in the center, but off-center instruments are hard to pinpoint.
Spaced-pair miking is a good choice if you want the sonic images to
be diffuse or blended, instead of sharply focused.
Another flaw of spaced mics: If you mix both mics to mono, you
may get phase cancellations of various frequencies. This may or may
not be audible.
Spaced mics, however, give a “warm” sense of ambience, in which
the concert-hall reverb seems to surround the instruments and,
sometimes, the listener. Here’s why: The two channels of recorded
reverb are incoherent; that is, they have random phase relationships.
Incoherent signals from stereo speakers sound diffuse and spacious.
Because spaced mics pick up reverb incoherently, it sounds diffuse
and spacious. The simulated spaciousness caused by this phasiness
isn’t necessarily realistic, but it’s pleasant to many listeners.
Another advantage of the spaced pair is that you can use omni mics.
An omni condenser mic has deeper bass than a uni condenser mic.
NEAR-COINCIDENT PAIR
In this method, you angle apart two directional mics, and space their
grilles a few inches apart horizontally (Figure 7.14). Even a few inches
of spacing increases the stereo spread and adds a sense of ambient warmth or air to the recording. The greater the angle or spacing
between mics, the greater the stereo spread. If the angle is too wide,
center images will be weak (there will be a “hole in the middle”).
Microphone Technique Basics CHAPTER 7
MUSICAL
ENSEMBLE
L
ANGLE
R
SPACING
RIGHT CHANNEL
OUTPUT
LEFT CHANNEL
OUTPUT
FIGURE 7.14
Near-coincident pair technique.
How does this method work? Angling directional mics produces
level differences between channels. Spacing mics produces time differences. The level differences and time differences combine to create
the stereo effect.
If the angling or spacing is too great, you get exaggerated separation.
If the angling or spacing is too small, you’ll hear a narrow stereo
spread.
A common near-coincident method is the ORTF system, which uses
two cardioids angled 110 degrees apart and spaced 7 inches (17 cm)
horizontally. Usually this method gives accurate localization; that is,
instruments at the sides of the orchestra are reproduced at or very
near the speakers, and instruments halfway to one side are reproduced about halfway to one side.
BAFFLED OMNI PAIR
This method uses two omni mics, usually ear-spaced, and separated
by either a hard or soft baffle (Figure 7.15). To create stereo, it uses
time differences at low frequencies and level differences at high frequencies. The spacing between mics creates time differences. The
baffle creates a sound shadow (reduced high frequencies) at the mic
farthest from the source. Between the two channels, there are spectral
differences—differences in frequency response.
Some examples of baffled-omni pairs are the Schoeps or Neumann
sphere microphones, the Jecklin Disk, and the Crown SASS-P MKII
stereo microphone.
119
120
CHAPTER 7 Microphone Technique Basics
MUSICAL
ENSEMBLE
EAR
SPACING
LEFT CHANNEL
OUTPUT
RIGHT CHANNEL
OUTPUT
FIGURE 7.15
Baffled-omni technique.
Comparing the Four Techniques
Coincident pair:
■
■
■
■
■
Uses two directional mics angled apart with grilles touching.
Level differences between channels produce the stereo effect.
Images are sharp.
Stereo spread ranges from narrow to accurate.
Signals are mono-compatible.
Spaced pair:
■
■
■
■
■
■
■
Uses two mics spaced several feet apart, aiming straight ahead.
Time differences between channels produce the stereo effect.
Off-center images are diffuse.
Stereo spread tends to be exaggerated unless a third center mic
is used, or unless spacing is under 2 to 3 feet.
Provides a warm sense of ambience.
Tends not to be mono-compatible, but there are exceptions.
Good low-frequency response if you use omni condensers.
Near-coincident pair:
■
Uses two directional mics angled apart and spaced a few inches
apart horizontally.
Microphone Technique Basics CHAPTER 7
■
■
■
■
Level and time differences between channels produce the stereo
effect.
Images are sharp.
Stereo spread tends to be accurate. Provides a greater sense of
air than coincident methods.
Tends not to be mono-compatible.
Play audio clip 22 at www.elsevierdirect.com/companions/9780240811444
to hear a comparison of the coincident, near-coincident and spaced-pair
techniques.
Baffled omni pair:
■
■
■
■
■
■
■
■
Uses two omni mics, usually ear-spaced, with a baffle between
them.
Level, time, and spectral differences produce the stereo effect.
Images are sharp.
Stereo spread tends to be accurate but is not adjustable (except
partly by panning).
Good low-frequency response.
Good imaging with headphones.
Provides more air than coincident methods.
Tends not to be mono-compatible, but there are exceptions.
HARDWARE
A handy device is a stereo mic adapter or stereo bar (Figure 7.16). It
mounts two mics on a single stand, and lets you adjust the angle and
spacing. You might prefer to use a stereo mic instead of two mics. It
has two mic capsules in a single housing for convenience.
HOW TO TEST IMAGING
Here’s a way to check the stereo imaging of a mic technique.
1. Set up the stereo mic array in front of a stage.
2. Record yourself speaking from various locations on stage where
the instruments will be—center, half-right, far right, half-left,
far left. Announce your position.
3. Play back the recording over speakers.
121
122
CHAPTER 7 Microphone Technique Basics
FIGURE 7.16
Stereo mic adapter.
You’ll hear how accurately the technique translated your positions,
and you’ll hear how sharp the images are.
We looked at several mic arrays to record in stereo. Each has its pros
and cons. Which method you choose depends on the sonic compromises you’re willing to make.
CHAPTER 8
Microphone Techniques
This chapter describes some ways to select and place mics for musical instruments and vocals. These techniques are popular, but they’re
just suggested starting points. Feel free to experiment.
Before you mike an instrument, listen to it live in the studio, so you
know what sound you’re starting with. You might want to duplicate
that sound through your monitor speakers.
ELECTRIC GUITAR
Let’s start by looking at the chain of guitar, effects, amplifier, and
speaker. At each point in the chain where you record, you’ll get a different sound (Figure 8.1).
1. The electric guitar puts out an electrical signal that sounds
clean and clear.
2. This signal might go through some effects boxes, such as distortion, wah wah, compression, chorus, or stereo effects.
3. Then the signal goes through a guitar amp, which boosts the
signal and adds distortion. At the amplifier output (preamp out
or external speaker jack), the sound is very bright and edgy.
4. The distorted amp signal is played by the speaker in the amp.
Because the speaker rolls off above 4 kHz, it takes the edge off
the distortion and makes it more pleasant.
You can record the electric guitar in many ways (Figure 8.1):
■
■
With a mic in front of the guitar amp
With a direct box
123
124
CHAPTER 8 Microphone Techniques
RECORD
HERE
RECORD
HERE
RECORD
HERE
FIGURE 8.1
Three places to record the electric guitar.
■
■
Both miked and direct
Through a signal processor or stomp box
The song you’re recording will tell you what method it wants. Just
mike the amp when you want a rough, raw sound with tube distortion and speaker coloration. Rock ‘n’ roll or heavy metal usually
sounds best with a miked amp. If you record through a direct box,
the sound is clean and clear, with crisp highs and deep lows. That
might work for quiet jazz or R&B. Use whatever sounds right for the
particular song you’re recording.
First, try to kill any hum you hear from the guitar amp. Turn up the
guitar’s volume and treble controls so that the guitar signal overrides
hum and noise picked up by the guitar cable. Ask the guitarist to
move around, or rotate, to find a spot in the room where hum disappears. Flip the polarity switch on the amp to the lowest hum position. To remove buzzes between guitar notes, try a noise gate, or ask
the player to keep his hands on the strings.
Miking the Amp
Small practice amplifiers tend to be better for recording than large,
noisy stage amps. If you use a small one, place it on a chair to avoid
picking up sound reflections from the floor (unless you like that
effect).
A common mic for the guitar amp is a cardioid dynamic type with
a “presence peak” in its frequency response (a boost around 5 kHz).
Microphone Techniques CHAPTER 8
The cardioid pattern reduces leakage (off-mic sounds from other
instruments). The dynamic type handles loud sounds without distorting, and the presence peak adds “bite.” Of course, you can use
any mic that sounds good to you.
As a starting point, try miking the amp about an inch from the grille
near a speaker cone, slightly off-center—where the cone meets the
dome.
The closer you mike the amp, the bassier the tone. The farther offcenter the mic is, the duller the tone. Often, distant miking sounds
great when you overdub a lead guitar solo played through a stack of
speakers in a live room. Try a boundary mic on the floor or on the
wall several feet away.
Recording Direct
Now let’s look at recording direct (also known as direct injection or
DI). The electric guitar produces an electrical signal that you can plug
into your mixer. You bypass the mic and guitar amp, so the sound
is clean and clear. Just remember that amp distortion is desirable in
some songs.
Mixer mic inputs tend to have an impedance (Z) around 1500 ohms.
But a guitar pickup is several thousand ohms. So if you plug a high-Z
electric guitar directly into a mic input, the input will load down the
pickup and give a thin or dull sound.
To get around this loading problem, use a direct box between the guitar and your mixer (Figure 8.2). The DI box has a high-Z input and
low-Z output, thanks to a built-in transformer or circuit. Some mixers
and audio interfaces have a high-Z (instrument) input jack built in,
so you can plug the electric guitar or bass directly into this jack.
The direct box should have a ground-lift switch to prevent ground
loops and hum. Set it to the position where you monitor the least
hum. You might try a mix of direct sound and miked sound. Play audio
clip 23 at www.elsevierdirect.com/companions/9780240811444 to hear demonstrations of electric-guitar recording methods.
It’s a good idea to record the guitar direct on its own track even if
you mike the amp. Later during mixdown, you can run the DI track
through a guitar-amp simulator plug-in, which might sound better
than the real guitar amp did.
125
126
CHAPTER 8 Microphone Techniques
INSTRUMENT
AMP
GND
LIFT
MIC-LEVEL
OUTPUT
FIGURE 8.2
Typical direct box.
Electric Guitar Effects
If you want to record the guitarist’s effects, connect the output of the
effects boxes into the direct-box input. Many players have a rack of
signal processors that creates their unique sound, and they just give
you their direct feed. Be open to their suggestions, and be diplomatic
about changing the sound. If they are studio players, they often have
a better handle on effects than you might as the engineer.
You might want a “fat” or spacious lead-guitar sound. Here are some
ways to get it:
■
■
■
■
■
Send the guitar signal through a digital delay set to 20 to 30 msec.
Pan guitar left, delay right. Adjust levels for nearly equal loudness
from each speaker. (Watch out for phase cancellations in mono.)
Send the guitar signal through a pitch-shifter, set for about
10 cents of pitch bending. Pan guitar left, pitch-shifted guitar
right. (A cent is 1/100 of an equal-temperament semitone. There
are 100 cents in a half-tone or semitone interval of pitch).
Record two guitarists playing identical parts, and pan them left
and right. This works great for rhythm-guitar parts in heavy
metal.
Double the guitar. Have the player re-record the same part on
an unused track while listening to the original part. Pan the
original part left and pan the new part right.
Add stereo reverb or stereo chorus.
Some guitar processors add many effects to an electric guitar, such as
distortion, EQ, chorus, and compression. An example is the Line 6 Pod.
Microphone Techniques CHAPTER 8
You simply plug the electric guitar into the processor, adjust it for the
desired sound, and record the signal direct. You wind up with a fully
produced sound with a minimum of effort.
Re-amping is a technique that lets you work on the amp’s sound during mixdown rather than during recording. Record the guitar direct,
then feed that track’s signal into a guitar processor or miked guitar
amp during mixdown. Use a low- to high-Z transformer between the
track output and the processor or amp input. Record the processor
or amp on an open track. In a digital audio workstation (DAW), you
can start with a track of a direct-recorded guitar, then insert a guitaramp modeling plug-in.
ELECTRIC BASS
BWAM, dik diddy bum. Do your bass tracks sound that clear? Or
are they more muddy, like, “Bwuh, dip dubba duh”? Here’s how to
record the electric bass so it’s clean and easy to hear in a mix.
As always, first you work on the sound of the instrument itself. Put
on new strings if the old ones sound dull. Adjust the pickup screws
(if any) for equal output from each string. Also adjust the intonation
and tuning.
Usually, you record the electric bass direct for the cleanest possible
sound. A direct pickup gives deeper lows than a miked amp, but the
amp gives more midrange punch. You might want to mix the direct
and miked sound. Use a condenser or dynamic mic with a good lowfrequency response, placed 1 to 6 inches from the speaker.
When mixing a direct signal and a mic signal, make sure they are
in-phase with each other. To do this, set them to equal levels and
reverse the polarity of the direct signal or the mic signal. The polarity
that gives the most bass is correct.
Have the musician play some scales to see if any notes are louder
than the rest. You might set a parametric equalizer to soften these
notes, or use a compressor.
The bass guitar should be fairly constant in level (a dynamic range of
about 6 dB) to be audible throughout the song, and to avoid clipping
the recording on loud peaks. To do this, run the bass guitar through
a compressor. Set the compression ratio to about 4:1; set the attack
127
128
CHAPTER 8 Microphone Techniques
time fairly slow (8 to 20 msec) to preserve the attack transient; and
set the release time fairly fast (1/2 second). If the release time is too
fast, you get harmonic distortion.
EQ can make the bass guitar clearer. Try cutting around 60 to 80 Hz,
or at 400 Hz. A boost at 2 to 2.5 kHz adds edge or slap, and a boost
at 700 to 900 Hz adds “growl” and harmonic clarity. If you boost the
lows around 100 Hz, try boosting at a lower frequency in the kick
drum’s EQ to keep their sounds distinct. A fretless bass will probably
need different EQ or less EQ than a fretted bass.
Here are some ways to make the bass sound clean and well defined:
■
■
■
■
■
■
■
■
Record the bass direct.
Use no reverb or echo on the bass.
Have the bass player turn down the bass amp in the studio, just
loud enough to play adequately. This reduces muddy-sounding
bass leakage into other mics.
Better yet, don’t use the amp. Instead, have the musicians monitor the bass (and each other) with headphones.
Have the bass player try new strings or a different bass. Some
basses are better for recording than others. Use roundwound
strings for a bright tone or flatwounds for a rounder tone.
Ask the bass player to use the treble pickup near the bridge.
Be sure to record the bass with enough edge or harmonics so
the bass will be audible on small, cheap speakers.
Try a bass-guitar signal processor such as the Line 6 Bass Pod,
Zoom B1, or DigiTech BP80.
If the bass part is full and sustained, it’s probably best to go for a
mellow sound without much pluck. Let the kick drum define the
rhythmic pattern. But if both the bass and kick are rhythmic and
work independently, then you should hear the plucks. Listen to the
song first, then get a bass sound appropriate for the music. A sharp,
twangy timbre is seldom right for a ballad; a full, round tone will get
lost in a fusion piece.
Often, a musician plays bass lines on a synth or sound module. The
module is triggered from a keyboard, a sequencer, or a bass guitar
plugged into a pitch-to-MIDI converter. Connect the module output
to your mixer or audio interface line input.
Two effects boxes for the electric bass are the octave box and the bass
chorus. The octave box takes the bass signal and drops it an octave
Microphone Techniques CHAPTER 8
in pitch; that is, it divides the bass signal’s fundamental frequency
in two. You put 82 Hz in; you get 41 Hz out. This gives an extra deep,
growly sound. So does a 5-string bass.
A bass chorus gives a wavy, shimmering effect. Like a conventional
chorus box, it detunes the signal and combines the detuned signal
with the direct signal. Also, it removes the lowest frequencies from the
detuned signal, so that the chorus effect doesn’t thin out the sound.
SYNTHESIZER, DRUM MACHINE, AND
ELECTRIC PIANO
For the most clarity, you usually DI a synth, MIDI sound module,
drum machine, or electric piano. Set the volume on the instrument
about three-quarters up to get a strong signal. Try to get the sound
you want from patch settings rather than EQ.
Plug the instrument into a phone jack input on your mixer, or use a
direct box. If you connect to a phone jack and hear hum, you probably have a ground loop. Here are some fixes:
■
■
■
Power your mixer and the instrument from the same outlet
strip. If necessary, use a thick extension cord between the outlet
strip and the instrument.
Use a direct box instead of a guitar cord, and set the ground-lift
switch to the position where you monitor the least hum.
To reduce hum from a low-cost synth, use battery power instead
of an AC adapter.
A synth can sound dry and sterile. To get a livelier, funkier sound,
you might run the synth signal into a power amp and speakers, and
mike the speakers a few feet away.
If the keyboard player has several keyboards plugged into a keyboard
mixer, you may want to record a premixed signal from that mixer’s
output. Record both outputs of stereo keyboards.
LESLIE ORGAN SPEAKER
This glorious device has a rotating dual-horn on top for highs and
a woofer on the bottom for lows. Only one horn of the two makes
sound; the other is for weight balance. The swirling, grungy sound
comes from the phasiness and Doppler effect of the rotating horn,
129
130
CHAPTER 8 Microphone Techniques
ONE MIC,
STEREO MIC,
OR MIC PAIR
WOOFER MIC
FIGURE 8.3
Miking a Leslie organ speaker.
and from the distorted tube electronics that drive the speaker. Here
are a few ways to record it (Figure 8.3):
■
■
In mono: Mike the top and bottom separately, 3 inches to 1
foot away. Aim the mics into the louvers. In the top mic’s signal, roll off the lows below 150 Hz.
In stereo: Record the top horn with a stereo mic or a pair of
mics out front. Put a mic with a good low end on the bottom
speaker, and pan it to center.
When you record the Leslie, watch out for wind noise from the rotating horn and buzz from the motor. Mike farther away if you monitor
these noises.
Rather than recording an actual Hammond B3 organ and Leslie
speaker, you might prefer to use a software emulation of those
instruments: an organ soft synth or sample and a Leslie speaker
plug-in. Trigger the synth or sample by a MIDI sequencer or MIDI
controller (covered in Chapter 16). You can automate the horn rotation speed in the Leslie speaker plug-in.
DRUM SET
The first step is to make the drums sound good live in the studio. If
the set sounds poor, you’ll have a hard time making it sound great in
Microphone Techniques CHAPTER 8
the control room! You might put the drum set on a riser 1–1/2 feet
high to reduce bass leakage and to provide better eye contact between
the drummer and the rest of the band. To reduce drum leakage into
other mics, you could surround the set with goboes—padded thickwood panels about 4 feet tall. For more isolation, place the set in a
drum booth, a small padded room with windows. It’s also common
to overdub the set in a live room.
Tuning
One secret to creating a good drum sound lies in careful tuning. It’s
easier to record a killer sound if you tune the set to sound right in
the studio before miking it.
First let’s consider drum heads. Plain heads have the most ring or
sustain, while heads with sound dots or hydraulic heads dampen the
ring. Thin heads are best for recording because they have crisp attack
and long sustain. Old heads become dull, so use new heads.
When you tune the toms, first take off the heads and remove the
damping mechanism, which can rattle. Put just the top head on and
hand-tighten the lugs. Then, using a drum key, tighten opposite pairs
of lugs one at a time, one full turn. After you tighten all the lugs,
repeat the process, tightening one-half turn. Then press on the head
to stretch it. Continue tightening a half-turn at a time until you reach
the pitch you want. You’ll get the most pleasing tone when the heads
are tuned within the range of the shell resonance.
To reduce ugly overtones, try to keep the tension the same around
the head. While touching the center of the head, tap with a drumstick on the head near each lug. Adjust tension for equal pitch
around the drum. If you want a downward pitch bend after the head
is struck, loosen one lug.
Keep the bottom head off the drum for the most projection and the
broadest range of tuning. In this case, pack the bottom lugs with
felt to prevent rattles. But you may want to add the bottom head for
extra control of the sound. Projection is best if the bottom head is
tighter than the top head—say, tuned a fourth above the top head.
There will be a muted attack, an “open” tone, and some note bending. If you tune the bottom head looser than the top, the tone will
be more “closed,” with good attack.
131
132
CHAPTER 8 Microphone Techniques
With the kick drum (bass drum), a loose head gives lots of slap and
attack, and almost no tone. The opposite is true for a tight head.
Tune the head to complement the style of music. For more attack or
click, use a hard beater.
Tune the snare drum with the snares off. A loose batter head or top
head gives a deep, fat sound. A tight batter head sounds bright and
crisp. With the snare head or bottom head loose, the tone is deep
with little snare buzz, while a tight snare head yields a crisp snare
response. Set the snare tension just to the point where the snare
wires begin to “choke” the sound, then back off a little.
Damping and Noise Control
Usually the heads should ring without any damping. But if the toms or
snare drum rings too much, put some plastic damping rings on them.
Or tape some gauze pads, tissues, or folded handkerchiefs to the edge
of the heads. Put masking tape on three sides of the pad so that the
untaped edge is free to vibrate and dampen the head motion. Don’t
overdo the damping, or the drum set will sound like cardboard boxes.
Oil the kick drum pedal to prevent squeaks. Tape rattling hardware
in place.
Sometimes a snare drum buzzes in sympathetic vibration with
a bass-guitar passage or a tom-tom fill. Try to control the buzz by
wedging a thick cotton wad between the snares and the drum stand.
Or tune the snare to a different pitch than the toms.
Drum Miking
Now you’re ready to mike the set. For a tight sound, place a mic near
each drum head. For a more open, airy sound, use fewer mics or mix
in some room mics placed several feet away. Typical room mics are
omni condensers or boundary mics. Figure 8.4 shows typical mic
placements for a rock drum set. Let’s look at each part of the kit.
SNARE
The most popular type of mic for the snare is a cardioid dynamic
with a presence peak. The cardioid pattern reduces leakage; its proximity effect boosts the bass for a fatter sound. The presence peak
adds attack. You might prefer a cardioid condenser for its sharp transient response.
Microphone Techniques CHAPTER 8
OVERHEAD
MIC
ALTERNATE MIKING
WITH A STEREO PAIR
OVERHEAD
MIC
RACK TOM MICS
SNARE
MIC
FLOOR TOM
MIC
KICK
MIC
FIGURE 8.4
Typical mic placements for a rock drum set.
FIGURE 8.5
Snare-drum miking.
Bring the mic in from the front of the set on a boom. Place the mic even
with the rim, 2 inches above the head (Figure 8.5). Angle the mic down
to aim where the drummer hits, or attach a mini condenser mic to the
side of the snare drum so it “looks at” the top head over the rim.
Some engineers mike both the top and bottom heads of the snare
drum, with the microphones in opposite polarity. A mic under the
133
134
CHAPTER 8 Microphone Techniques
snare drum gives a zippy sound; a mic over the snare drum gives
a fuller sound. You might prefer to use just a top mic, and move it
around until it picks up both the top head and snares. The sound
is full with the mic near the top head, and thins out and becomes
brighter as you move the mic toward the rim and down the side of
the drum.
Whenever the hi-hat closes, it makes a puff of air that can “pop” the
snare-drum mic. Place the snare mic so the air puff doesn’t hit it. To
prevent hi-hat leakage into the snare mic:
■
■
■
■
■
Mike the snare closely.
Bring the snare boom in under the hi-hat, and aim the snare
mic away from the hi-hat.
Use a piece of foam or pillow to block sound from the hi-hat.
Use a de-esser on the snare.
Don’t play the hi-hat during tracking—overdub it later.
HI-HAT
Try a cardioid condenser mic about 6 inches over the cymbal edge
that’s farthest from the drummer (Figure 8.6). To avoid the air puff
FIGURE 8.6
Hi-hat miking.
Microphone Techniques CHAPTER 8
just mentioned, don’t mike the hi-hat off its side; mike it from
above aiming down. This also reduces snare leakage. Filter out the
lows below about 500 Hz. You may not need a hi-hat microphone,
especially if you use room mics. Usually the overhead mics pick up
enough hi-hat.
TOM-TOMS
You can mike the toms individually, or put a mic between each
pair of toms. The first option sounds more bassy. Place a cardioid
dynamic about 2 inches over the drumhead and 1 inch in from the
rim, angled down about 45 degrees toward the head (Figure 8.7).
Again, the cardioid’s proximity effect gives a full sound. Another way
is to clip mini condenser mics to the toms, peeking over the top rim
of each drum.
If the tom mics pick up too much of the cymbals, aim the “dead”
rear of the tom mics at the cymbals. If you use a supercardioid or
hypercardioid mic, aim the null of best rejection at the cymbals.
FIGURE 8.7
Tom-tom miking.
135
136
CHAPTER 8 Microphone Techniques
KICK DRUM
Place a blanket or folded towel inside the drum, pressing against the
beater head to dampen the vibration and tighten the beat. The blanket shortens the decay portion of the kick-drum envelope. To emphasize the attack, use a wood or plastic beater—not felt—and tune the
drum low.
A popular mic for kick drum is a large-diameter, cardioid dynamic
type with an extended low-frequency response. Some mics are
designed specifically for the kick drum, such as the AKG D112, AudioTechnica AT AE2500, Electro-Voice N/D868, and Shure Beta 52A.
For starters, place the kick mic inside on a boom, a few inches from
where the beater hits (Figure 8.8). Mic placement close to the beater
picks up a hard beater sound; off-center placement picks up more
skin tone, and farther away picks up a boomier shell sound.
How should the recorded kick drum sound? Well, they don’t call it
a kick drum for nothing. THUNK! You should hear a powerful lowend thump plus an attack transient.
FIGURE 8.8
Kick-drum miking.
Microphone Techniques CHAPTER 8
CYMBALS
To capture all the crisp “ping” of the cymbals, a good mic choice is a
cardioid condenser with an extended high-frequency response, flat or
rising at high frequencies. Place the overhead mics about 2 to 3 feet
above the cymbal edges; closer miking picks up a low-frequency ring.
The cymbal edges radiate the most highs. Place the cymbal mics to pick
up all the cymbals equally. Try to place them the same distance from
the snare. If your recording will be heard in mono, or for sharper imaging, you might want to mount the mic grilles together and angle the
mics apart (Figure 8.4). This results in a narrow stereo spread. Another
option is a stereo mic overhead. For wide stereo and sharp imaging, try
a near-coincident pair aimed at the high-hat and floor tom.
Recorded cymbals should sound crisp and smooth, not muffled
or harsh.
ROOM MICS
Besides the close-up drum mics, you might want to use a distant pair
of room mics when you record drum overdubs. Place the mics about
10 or 20 feet from the set to pick up room reverb. When mixed with
the close-up mics, the room mics give an open, airy sound to the
drums. Popular room mics are omni condensers or boundary mics
taped to the control-room window. You might compress the room
mics for special effect. If you don’t have enough tracks for room
microphones, try raising the overhead mics.
BOUNDARY MIC TECHNIQUES
Boundary mics let you pick up the set in unusual ways. You can strap
one on the drummer’s chest to pick up the set as the drummer hears
it. Tape them to hard-surfaced goboes surrounding the drummer.
Put them on the floor under the toms and near the kick drum, or
hang a pair over the cymbals. Try a supercardioid boundary mic in
the kick drum.
RECORDING WITH TWO TO FOUR MICS
Sometimes you can mike the set simply. Place a stereo mic (or
two mics) overhead and put another mic in the kick. If necessary,
137
138
CHAPTER 8 Microphone Techniques
OVERHEAD
MIC
ALTERNATE MIKING
WITH A STEREO PAIR
OVERHEAD
MIC
SNARE
MIC
KICK
MIC
FIGURE 8.9
Miking a drum set with four mics.
add a snare-drum mic (Figure 8.9). This method works well for
acoustic jazz, and often for rock. If you want the toms to sound
fuller, boost the lows in the overhead mics. As an alternative try two
mics about 18 inches apart angled down at the set from just over the
drummer’s head.
Another setup is shown in Figure 8.10. It uses only one mini omni
condenser mic and one kick-drum mic. This method sometimes
works well on small drum sets. Clip a mini omni condenser mic to
the snare-drum rim about 4 inches above the rim, in the center of
the set, aiming at the hi-hat. Or place a small-diaphragm omni there
with a mic boom. Also mike the kick drum.
The mini mic will pick up the snare, hi-hat, and toms all around it,
and will pick up the cymbals from underneath. Move the mic closer
or farther from the toms, and raise or lower the cymbals, until you
hear a pleasing balance. Add a little bass and treble. You’ll be surprised at the good sound and even coverage you can get with this
simple setup.
Microphone Techniques CHAPTER 8
MINI OMNI
MIC
KICK
MIC
FIGURE 8.10
Miking a drum set with a mini omni mic.
MINI OMNI
MIC
MINI OMNI
MIC
KICK
MIC
FIGURE 8.11
Miking a drum set with two mini omni mics.
Want a stereo effect? Mount one mic 4 inches above the snare drum
rim between the hi-hat, snare drum, and rack tom. Adjust position
for best balance. Mount another mic four inches above the floor-tom
rim, on the side farthest from the drummer (Figure 8.11). Pan the
mini mics left and right.
Audio clip 24 at www.elsevierdirect.com/companions/9780240811444
demonstrates several methods of miking a drum set.
139
140
CHAPTER 8 Microphone Techniques
Drum Recording Tips
After you set up all the mics, ask the drummer to play. Listen for rattles and leakage by soloing each microphone. Try not to spend much
time getting a sound; otherwise you waste the other musicians’ time
and wear out the drummer.
To keep the drum sound tight during mixdown, mute or delete drum
tracks that aren’t in use in a particular tune, use a noise gate on the
kick and toms, or overdub the drums.
One dated effect for the snare drum is gated reverb. It’s a short splash
of bright-sounding reverberation, which is rapidly cut off by a noise
gate or expander. Many effects units have a gated-reverb program.
Another trick is recording “hot.” Using an analog multitrack (or its
plug-in equivalent), record the drums at a high level so they distort
just a little. It’s also common to compress the kick and snare.
A drummer might use drum pads, or drum triggers, fed into a sound
module. Record directly off the module. You might want to mike the
cymbals anyway for best sound.
If you’re recording a drum machine and it sounds too mechanical,
add some real drums. The machine can play a steady background
while the drummer plays fills.
When miking drums on stage for PA, you don’t need a forest of
unsightly mic stands and booms. Instead, you can use short mic
holders that clip onto drum rims and cymbal stands (Mic-Eze at
www.ac-cetera.com), or use mini condenser mics.
In a typical rock mix, the drums either are the loudest element, or
are slightly quieter than the lead vocal. The kick drum is almost as
loud as the snare. If you don’t want a wimpy mix, keep those drums
up front!
Try these EQ settings to enhance the recorded sound of the drums:
■
Snare: Fat at 200 Hz, crack at 5 kHz, sizzle at 10 or 12 kHz. Some
snare drums ring a lot at one note. To fix it, set up an equalizer on the snare-drum track (equalization, EQ, is explained
in Chapter 10.) Set a narrow, high-Q boost around 500 Hz and
sweep it in frequency until you amplify the frequency that is
ringing. Apply a narrow, high-Q cut there to remove the ringing.
Microphone Techniques CHAPTER 8
■
■
■
Toms: Cut around 600 to 800 Hz to reduce the papery sound.
Then if necessary, boost around 100 to 200 Hz for more fullness on rack toms or 80 to 100 Hz on floor toms. Boost around
5 kHz for more attack.
Cymbals: Sizzle at 10 kHz or higher. If you’re close-miking the
toms, you might roll off the lows in the cymbal mics below
500 Hz to reduce low-frequency leakage.
Kick drum: To remove the “cardboard” sound, cut at 300 to
600 Hz. Then if necessary, boost at 3 to 5 kHz for more click.
Don’t overdo the high-frequency boost; usually you don’t want
too much “point” on the kick sound. Boost 60 to 80 Hz if the
kick sounds thin. Filter out highs above 9 kHz to reduce leakage
from cymbals.
Try these tricks to come up with unusual drum sounds:
■
■
■
■
■
■
Record with a cheap dynamic or crystal mic, maybe in a can.
Run the drums through extreme processing: compression, gating, distortion, pitch shifting, tremolo, and so on.
Substitute other objects for drums, cymbals, drumsticks, and
brushes.
Move a mic around a cymbal or drumhead while recording it.
Put the drums in a reverberant room or hallway.
Try the preverb effect described in Chapter 10.
Instead of recording an acoustic drum set, you might use an electronic
drum set or CD of drum samples. Copy the samples into a sampler or
sampling software, then trigger them with a MIDI sequencer or MIDI
controller with drum pads.
PERCUSSION
Let’s move on to percussion, such as the cowbell, triangle, tambourine, or bell tree. A good mic for metal percussion is a condenser type
because it has sharp transient response. Mike at least 1 foot away so
the mic doesn’t distort.
You can pick up congas, bongos, and timbales with a single mic
between the pair, a few inches over the top rim, aimed at the heads.
Or put a mic on each drum. It often helps to mike these drums top
and bottom, with the bottom mic in opposite polarity. A cardioid
dynamic with a presence peak gives a full sound with a clear attack.
141
142
CHAPTER 8 Microphone Techniques
For xylophones and vibraphones, place two cardioid mics 1–1/2
feet above the instrument, aiming down. Cross the mics 135 degrees
apart or place them about 2 feet apart. You’ll get a balanced pickup
of the whole instrument.
ACOUSTIC GUITAR
The acoustic guitar has a delicate timbre that you can capture
through careful mic selection and placement. First prepare the acoustic guitar for recording. To reduce finger squeaks, try commercial
string lubricant, a household cleaner/waxer, talcum powder on fingers, or smooth-wound strings. Ask the guitarist to play louder; this
increases the “music-to-squeak” ratio!
Replace old strings with new ones a few days before the session.
Experiment with different kinds of guitars, picks, and finger picking
to get a sound that’s right for the song.
For acoustic guitar, a popular mic is a pencil-type condenser with a
smooth, extended frequency response from 80 Hz up. This kind of
mic has a clear, detailed sound. You can hear each string plucked in
a strummed chord. Usually the sound picked up is as crisp as the
real thing.
Now let’s look at some mic positions. To record a classical guitar solo
in a recital hall, mike about 3 to 6 feet away to pick up room reverb.
Try a stereo pair (Figure 8.12A), such as XY, ORTF, MS, or spaced pair
(described in Chapter 7). If you record a classical guitar solo in a
dead studio, mike about 1.5 to 2 feet away and add artificial reverb.
When you record pop, folk, or rock music, try a spot about 6 to 12
inches from where the fingerboard joins the guitar body (Figure
8.12B). That’s a good starting point for capturing the acoustic guitar
accurately. Still, you need to experiment and use your ears. Close to
the bridge, the sound is woody and mellow.
In general, close miking gives more isolation, but tends to sound
harsh and aggressive. Distant miking lets the instrument “breathe”;
you hear a gentler, more open sound.
Another spot to try: Tape a mini omni mic onto the body near the
bottom of the sound hole, and roll off the excess bass. This spot
gives good isolation (Figure 8.12C).
Microphone Techniques CHAPTER 8
NATURAL
B
STEREO MIKING
A CLASSICAL SOLO
A
STEREO
D
F
BASSY BUT
GOOD ISOLATION
E
C
MINI MIC
FIGURE 8.12
Some mic techniques for acoustic guitar.
The guitar will sound more real if you record in stereo. Try one mic
near the bridge, and another near the 12th fret (Figures 8.12D and E).
Pan part-way left and right. Another way to record stereo is with an
XY pair of cardioid mics about 6 inches from the 12th or 16th fret,
mixed with a 3-foot-spaced pair of omni mics about 3 feet away.
Is feedback or leakage a problem? Mike close to the sound hole
(Figure 8.12F). The tone there is very bassy, so turn down the lowfrequency EQ on your mixer until the sound is natural. Also cut a
few decibels around 3 kHz to reduce harshness.
You get the most isolation with a contact pickup. It attaches to the
guitar, usually under the bridge. The sound of a pickup is something
like an electric guitar. You can mix a mic with a pickup to add air
and string noise to the sound of the pickup. That way, you get good
isolation and good tone quality. You might record the acoustic guitar off its pickup while tracking to prevent leakage, then overdub the
guitar later with a microphone for its better sound quality.
SINGER/GUITARIST
Normally you overdub the guitar and vocal separately. But if you
have to record both at once, the vocal might sound filtered or hollow
because of phase cancellations between the vocal mic and guitar mic.
143
144
CHAPTER 8 Microphone Techniques
This can happen whenever two mics pick up the same source at
approximately equal levels, at different distances, and both mixed to
the same channel. Try one of these methods to solve the problem:
■
■
■
■
■
■
Angle the vocal mic up and angle the guitar mic down to isolate
the two sources. Follow the 3:1 rule described in Chapter 7.
Mike the voice and guitar very close. Roll off the excess bass
with your mixer’s EQ.
Use a pickup on the guitar instead of a mic.
Place two bidirectional mics so the tops of their grilles touch.
This gets rid of any delay between their signals. Aim the “dead”
side of the vocal mic at the guitar; aim the dead side of the guitar mic at the mouth.
Use just one mic, or a stereo mic, midway between the mouth
and guitar about 1 foot out front. Adjust the balance between
voice and guitar by changing the mic’s height.
Delay the vocal mic signal by about 1 msec. Then the signals of
the two mics will be more in-phase, preventing phase cancellations when they are mixed to the same channel. Some multitrack recorders have a track-delay feature for this purpose.
GRAND PIANO
This magnificent instrument is a challenge to record well. First have
the piano tuned, and oil the pedals to reduce squeaks. You can
prevent thumps by stuffing some foam or cloth under the pedal
mechanism.
For a classical-music solo, record in a reverberant room such as a
recital hall or concert hall. Reverb is part of the sound. Set the piano
lid on the long stick. Use condenser mics with a flat response. Place
a stereo mic, or a stereo pair of cardioid mics, about 7 feet away and
7 feet high, up to 9 feet away and 9 feet high (Figure 8.13). Move the
mics closer to reduce reverb, farther to increase it. When using a pair
of omni mics, place them 1.3 to 2 feet apart, 3 to 6.5 feet from the
piano, and 4 to 5 feet high (Figure 8.14). You might need to mix in
a pair of hall mics: try cardioids aiming away from the piano about
25 feet away.
When recording a piano concerto, give the piano a spot mic about
1 to 3 feet away. Put the mic in a shock mount.
Microphone Techniques CHAPTER 8
STEREO
MIC PAIR
7 TO 9
FT
7 TO 9
FT
STEREO
MIC PAIR
FIGURE 8.13
Suggested grand-piano miking for classical music (using cardioid mics).
Omni mics
3 ft. – 6.5 ft.
4 ft. – 5 ft.
Omni mics
1.3 ft. – 2 ft.
FIGURE 8.14
Suggested grand-piano miking for classical music (using omnidirectional mics).
Pop music demands close miking. Close mics pick up less room
acoustics and leakage, and give a clear sound that cuts through the
mix. Try not to mike the strings closer than 8 inches, or else you’ll
emphasize the strings closest to the mics. You want equal coverage of
all the notes the pianist plays.
One popular method uses two spaced mics inside the piano. Use
omni or cardioid condensers, ideally in shock mounts. Put the lid
on the long stick. If you can, remove the lid to reduce boominess.
Center one mic over the treble strings and one over the bass strings.
145
146
CHAPTER 8 Microphone Techniques
BASS MIC
BASS MIC
STEREO PAIR
(ALTERNATIVE)
TREBLE MIC
TREBLE MIC
HAMMERS
MICS angled or
straight down
HAMMERS
8 TO 12 IN.
8 TO 12 IN.
A
B
BASS MIC
TREBLE MIC
C
Boundary mics gaffer-taped
to underside of raised lid
FIGURE 8.15
Suggested grand-piano miking for popular music.
Typically, both mics are 8 to 12 inches over the strings and 8 inches
horizontally from the hammers (Figure 8.15A, bass and treble mics).
Aim the mics straight down or angle them to aim at the hammers.
Pan the mics partly left and right for stereo.
One alternative is to put the treble mic near the hammers, and put
the bass mic about 2 feet toward the tail (Figure 8.15B). Another
method uses two ear-spaced omni condensers or an ORTF pair about
12 to 18 inches above the strings. Audio clip 25 at www.elsevierdirect
.com/companions/9780240811444 demonstrates some mic techniques for
grand piano.
The spaced mics might have phase cancellations when mixed to
mono, so you might want to try coincident miking (Figure 8.15A,
Microphone Techniques CHAPTER 8
stereo pair). Boom-mount a stereo mic, or an XY pair of cardioids
crossed at 120 degrees. Miking close to the hammers sounds percussive; toward the tail has more tone.
For more clarity and attack, boost EQ around 10 kHz or use a mic
with a rising high-frequency response.
Boundary mics work well, too. If you want to pick up the piano in
mono, tape a boundary mic to the underside of the raised lid, in the
center of the strings, near the hammers. Use two for stereo over the
bass and treble strings. Put the bass mic near the tail of the piano to
equalize the mic distances to the hammers (Figure 8.15C). If leakage is a problem, close the lid and cut EQ a little around 250 Hz to
reduce boominess.
If your studio lacks a piano, consider using a software emulation of a
piano. Some programs provide high-quality samples of piano notes
that can be played with a sequencer or a MIDI controller. Examples
include: Steinberg Grand VST 2.0 ($199 at www.steinberg.net) and
Maxim Digital Audio Piano (freeware at www.mda-vst.com under
VST synths).
UPRIGHT PIANO
Here are some ways to mike an upright piano:
(Figure 8.16A) Remove the panel in front of the piano to expose the
strings over the keyboard. Place one mic near the bass strings and
one near the treble strings about 8 inches away. Record in stereo and
pan the signals left and right for the desired piano width. If you can
spare only one mic for the piano, just cover the treble strings.
(Not shown) Remove the top lid and upper panel. Put a stereo pair
of mics about 1 foot in front and 1 foot over the top. If the piano
is against a wall, angle the piano about 17 degrees from the wall to
reduce tubby resonances.
(Figure 8.16B) Aim the soundboard into the room. Mike the bass
and treble sides of the soundboard a few inches away. In this spot,
the mics pick up less pedal thumps and other noises. Try cardioid
dynamic mics with a presence peak.
147
148
CHAPTER 8 Microphone Techniques
PIANO
MICS
SOUND
BOARD
B
MIC
MIC
MICS
OPEN
A
FIGURE 8.16
Some mic techniques for upright piano.
ACOUSTIC BASS
The acoustic bass (string bass, double bass, upright bass) puts out
frequencies as low as 41 Hz, so use a mic with an extended lowfrequency response such as a large-diaphragm condenser mic or ribbon mic. As always, closer miking improves isolation, while distant
miking tends to sound more natural but can pick up too much room
sound. Try these techniques (Figure 8.17):
■
■
■
4 to 8 inches in front of the bridge, a few inches above the bridge.
4 to 6 inches under the bridge, a few inches from the strings.
This mic will pick up a deep sound with good definition. You
might mix in a second mic near the plucking fingers for clarity.
Mix a pickup with a mic, or use a pickup alone and EQ it to
sound good.
Here are some methods that isolate the bass and let the player move
around. They work well for PA:
■
■
Wrap a mini omni condenser mic in foam rubber (or in a foam
windscreen) and mount it in the bridge aiming up (Figure 8.17).
Tape a mini omni mic to the bridge, or wedge it into a slot in
the bridge.
Microphone Techniques CHAPTER 8
ACOUSTIC BASS
SIDE VIEW
NEAR PLUCKING
FINGERS
MIC OUT FRONT
4–8 IN.
MIC IN FOAM
BEHIND BRIDGE
NEAR BOTTOM AREA
FIGURE 8.17
Some mic techniques for the acoustic bass.
■
■
Wrap a regular cardioid mic in foam padding (except the front
grille) and squeeze it behind the bridge (Figure 8.17) or tailpiece.
For best isolation, try a direct feed from a pickup. This method
adds clarity and deep bass, but probably will need some EQ.
You might mix the pickup with a microphone.
BANJO
Try a flat-response condenser mic about 1 foot away from the center
(Figure 8.18). Or try a cardioid dynamic mic 6 inches from the lower
right area or 6 inches from the lower rim, aiming at the bridge. If you
need more isolation, mike closer and roll off some bass. The banjo
sounds pleasantly mellow when miked toward the edge of the head,
near the resonator holes (if the banjo has them). Cloth stuffed inside
will tighten the sound and will reduce feedback in PA situations.
For the most isolation, tape a mini omni condenser mic to the head
about 1 inch in from the bottom edge, or on the tailpiece, or on the
bridge. You can wedge a pickup between the strings below the bridge
and the banjo head. Put the pickup flat against the head surface.
149
150
CHAPTER 8 Microphone Techniques
BANJO
MINI OMNI
ON BRIDGE
MIC 1 FOOT
OUT FRONT
MINI OMNI
ON HEAD
MIC 6 INCHES
FROM LOWERRIGHT AREA
FIGURE 8.18
Four methods for miking a banjo.
MANDOLIN, DOBRO, BOUZOUKI, AND LAP
DULCIMER
Mike these about 6 to 8 inches away from the hole with a small-diaphragm condenser mic. Some engineers also mix in a mic near the
neck or dobro resonator. If you need more lows and more isolation,
tape a mini omni condenser mic near a hole and tweak EQ for the
best sound.
HAMMERED DULCIMER
Place a flat-response condenser mic about 2 feet over the center of
the soundboard (Figure 8.19A). On stage, place a cardioid dynamic
or condenser 6 to 12 inches over the middle of the top end (Figure
8.19B). For the best gain-before-feedback in a PA system, mix in a
mini omni condenser mic (or a cardioid with bass rolloff) very near
the sound hole (Figure 8.19C).
FIDDLE (VIOLIN)
Listen to the fiddle itself to make sure it sounds good. Correct any
instrument problems before miking.
First try a flat-response condenser mic (omni or cardioid) about 2
feet over the bridge. This distant miking gives an airy, silky sound.
Close miking (about 6 to 12 inches, Figure 8.20) sounds more
aggressive, which is desirable in old-time or bluegrass music. Aim the
Microphone Techniques CHAPTER 8
FIGURE 8.19
Some mic techniques for hammered dulcimer.
OVERHEAD
FOR SINGING
FIDDLER
SIDE
FIGURE 8.20
Three fiddle-miking methods.
mic toward the f-holes for warmth or toward the fingerboard for clarity. A fiddle that sounds too bright and scratchy can be tamed using a
ribbon mic, or by miking it from the side or even underneath. If the
ceiling over the fiddle is low, nail a square yard of acoustic foam up
there to prevent reflections.
If you have to mike close—say, for a singing fiddler—aim the mic
at the player’s chin from 6 to 12 inches above the end of the fingerboard. The mic will pick up both the singer and the fiddle.
If you need more isolation, try a mini omni mic. Wrap its cable
in foam rubber (or a windscreen) 1–1/2 inches from the capsule.
Wedge the foam under the tailpiece, and position the mic capsule
151
152
CHAPTER 8 Microphone Techniques
MINI OMNI
MIC PLACEMENTS
ON THE FIDDLE
OVER
F-HOLE,
CLIPPED TO
TAILPIECE
BETWEEN
TAILPIECE
AND BRIDGE
FIGURE 8.21
Two ways to close-mike a fiddle for isolation.
halfway between the tailpiece and bridge, a half inch over the body
(Figure 8.21). If necessary, cut a little at 3 kHz to reduce harshness
and boost around 200 Hz for warmth. Another option is to clip the
mic to the tailpiece and mount it over an f-hole.
A good spot for a pickup is on the left side of the top (player’s view),
on the player’s side of the bridge.
To record a classical violin solo, try a stereo mic (or a stereo pair) 5
to 15 feet away in a reverberant room.
STRING SECTION
Place the strings in a large, live room and mike them at a distance
to pick up a natural acoustic sound. A common mic choice is a condenser with a flat response. First try a stereo mic or stereo pair of mics
about 4 to 20 feet behind the conductor, raised about 13 to 15 feet.
If the room is noisy or too dead, or the balance is poor, you’ll need
to mike close and add digital reverb. Try one mic on every two to
four violins, 6 feet off the floor, aiming down. Same for the violas.
Mike the cello about 1 to 2 feet from the bridge, to the right side
between the bridge and f-hole. When you mix the strings to stereo,
pan them evenly between the monitor speakers. Spread them left,
center, and right to make a “curtain of sound.” If you can spare only
one track for the strings, use a stereoizer effect during mixdown.
Microphone Techniques CHAPTER 8
STRING QUARTET
Record a quartet in stereo using a stereo mic or a pair of mics.
Place them about 6 to 10 feet away to capture the room ambience.
The monitored instruments should not spread all the way between
speakers. If you want to narrow the stereo stage, angle or space the
mics closer together.
BLUEGRASS BAND AND OLD-TIME STRING BAND
Suppose you’re recording a group that has a good acoustic balance.
Try a stereo mic or stereo pair of mics about 2 to 3 feet away and 5
feet high (lower if the group is seated). Move the players toward or
away from the mics to adjust their balance.
You’ll have more control if you mike all the instruments up close
and mix them. This also gives a more “commercial” sound. The production style aims for a natural timbre on all the instruments, either
with no effects or with slight reverb.
HARP
Use a condenser mic with a flat response. If the harp is playing with
an orchestra, mike the harp about 18 inches from the front of the
soundboard, or 18 inches from the player’s left hand. With a Celtic
group, try a mic 1/3 of the way up the sound board about 6 to
12 inches away.
Tape a mini omni condenser mic to the soundboard if you need isolation. A mic on the inside of the soundboard has more isolation;
a mic on the outside sounds more natural. Also try a cardioid condenser wrapped in foam, stuck into the center hole from the rear.
HORNS
“Horns” in studio parlance refers to the brass instruments: trumpets,
cornets, trombones, baritones, French horns, and tubas.
All the brass radiate strong highs straight out from the bell, but do
not project them to the sides. A mic close to and in front of the bell
picks up a bright, edgy tone. To mellow out the tone, mike the bell
off-axis with a flat-response mic (Figure 8.22). The sound on-axis to
153
154
CHAPTER 8 Microphone Techniques
MELLOW
BRIGHT
TRUMPET – TOP VIEW
FIGURE 8.22
Miking for trumpet tone control.
the bell has a lot of spiky high harmonics that can overload a condenser mic, mixer input, or analog tape. That’s another reason to
mike off-axis.
Mike the trumpet with a dynamic or ribbon mic to take the edge
off the sound. Use a condenser mic if you want a lot of sizzle. Mike
about 1 foot away for a tight sound; mike several feet away for a
fuller, more dramatic sound.
You can pick up two or more horns with one microphone. Several
players can be grouped around a single omni mic, or around a stereo pair of mics. The musicians can play to a pair of boundary mics
taped on the control-room window or on a large panel.
Record a classical brass quartet in a reverberant room. Use a stereo
mic, or a stereo pair of mics, about 6 to 12 feet away.
SAXOPHONE
A sax miked very near the bell sounds bright, breathy, and rather
hard (Figure 8.23). Mike it there for best isolation. To get a warm,
natural sound, mike the sax about 1–1/2 feet away, halfway down
the wind column. Don’t mike too close, or else the level varies when
the player moves. A compromise position for a close-up mic is just
above the bell, aiming at the holes. You can group a sax section
around one mic.
Figure 8.24 shows a typical miking setup for big-band jazz. It uses
the techniques already described for the drums, bass, piano, electric
guitar, trumpet, and sax.
Microphone Techniques CHAPTER 8
FIGURE 8.23
Two ways to mike a saxophone.
TROMBONES
DRUMS
BASS
TRUMPETS
DI
GUITAR
SAXES
PIANO
MUSIC STANDS
FIGURE 8.24
Typical miking setup for big-band jazz.
WOODWINDS
With woodwinds, most of the sound radiates not from the bell, but
from the holes. So aim a flat-response mic at the holes about 1 foot
away (Figure 8.25).
When miking a woodwind section within an orchestra, you need to
reject nearby leakage from other instruments. To do that, try aiming
a bidirectional mic down over the woodwind section. The side nulls
of the mic cut down on leakage.
To pick up a flute in a pop-music group, try miking a few inches
from the area between the mouthpiece and the first set of finger
155
156
CHAPTER 8 Microphone Techniques
MIC
FIGURE 8.25
Miking a clarinet from the side.
FLUTE
TOP
VIEW
MIC
MINI OMNI
MIC
FRONT VIEW
FIGURE 8.26
Two methods of miking a flute.
holes (Figure 8.26). You may need a pop filter. If you want to reduce
breath noise, roll off high frequencies or mike farther away. You also
can attach a mini omni mic to the flute a few inches above the body,
between the mouthpiece and finger holes.
For classical music solos, try a stereo pair 4 to 12 feet away.
HARMONICA, ACCORDION, AND BAGPIPE
One way to mike a harmonica (harp) is to use a cardioid dynamic
mic with a ball grille. Place the mic very close to the harmonica or
have the player hold it. A condenser mic about 1 foot away gives a
Microphone Techniques CHAPTER 8
natural sound. To get a bluesy, dirty sound, use a “bullet”-type harmonica mic or play the harmonica through a miked guitar amp.
For an accordion, try a mic about 6 to 12 inches from the sound
holes near the keyboard. Some accordions have sound holes on
both sides, so you’ll need two mics. Follow the 3:1 rule. The distance
between mics should be at least three times the mic-to-source distance. One end of the accordion is in constant motion, so you might
want to attach a mini omni mic to that end. A solo accordion or
concertina could be miked with a stereo pair of flat-response cardioid condenser mics about 3 to 6 feet in front.
A bagpipe has two main sound sources: the chanter, which the musician plays with the fingers, and the drone pipes, which make a steady
tone. Mike the chanter about a foot away from the side, and mike
the drone pipes a foot from the end. Again, follow the 3:1 rule. You
could also mike the bagpipe a few feet away with one mic.
LEAD VOCAL
The lead vocal is the most important part of a pop song, so it’s critical to record it right. First set up a comfortable environment for the
singer. Put down a rug, add some flowers or candles, dim the lights.
Set up a good cue mix with effects to help the singer get into the
mood of the song.
You might want to turn off the reverb in the singer’s headphones;
this makes it easier to hear pitch. If the vocalist is singing flat, reduce
their headphone volume, and vice versa.
With any vocal recording, there are some problems to overcome,
but we can deal with them. Among these are proximity effect, breath
pops, wide dynamic range, sibilance, and sound reflections from the
music stand. Let’s look at these in detail.
Miking Distance
When you sing or talk close to most directional mics, the microphone boosts the bass in your voice. This is called the proximity
effect. We’ve come to accept this bassy sound as normal in a PA system, but the effect just sounds boomy in a recording.
157
158
CHAPTER 8 Microphone Techniques
FIGURE 8.27
Typical miking technique for a lead vocal.
To prevent boomy bass, mike the singer at a distance, about 8 inches
away (Figure 8.27). A popular mic choice is a flat-response condenser mic with a large diaphragm (1–1/4-inch diameter). As always,
you can use any mic that sounds good to you. If the mic has a bass
rolloff switch, set it to “flat.”
Singers should maintain their distance to the mic. I ask the singer
to spread the fingers, touch lips with the thumb, and touch the mic
with the pinky. The hand forms a spacer for keeping a constant
distance.
If the singer’s voice is too harsh or edgy, try miking partly to the side
of the mouth, aiming at the mouth. Or try a ribbon mic, or a multiband compressor set to compress from 2000 Hz and up.
Some singers feel more comfortable singing into a handheld mic.
You can give them a handheld mic but also mike them several inches
away with a good condenser microphone. Record both mics on separate tracks and choose whatever sounds the best.
If you must record the singer and the band at the same time—as in
a concert—you’ll have to mike close to avoid picking up the instruments with the vocal mic. Try a cardioid or supercardioid mic with a
foam pop filter. The sound will be bassy because of proximity effect,
so roll off the excess lows at your mixer. For starters, try ⫺6 dB at 100
to 200 Hz. Some mics have a bass filter switch for this purpose. Aim
the mic partly toward the singer’s nose to prevent a nasal or closednose effect. This close-up method works well if you want an intimate, breathy sound.
Microphone Techniques CHAPTER 8
When recording a classical-music singer who is accompanied by an
orchestra, place the mic about 1 to 2 feet away. If the singer is a soloist (maybe accompanied by piano), use a stereo pair about 8 to 15
feet away to pick up room reverb.
Breath Pops
When you sing a word with “p” or “t” sounds, a turbulent puff of air
shoots out of the mouth. The puff hits the mic and makes a thump
or small explosion called a pop. To reduce it, put a foam-plastic pop
filter on the mic. Some mics have a ball grille screen to cut pops, but
foam works better. The pop filter should be made of special opencell foam to pass high frequencies. For best pop rejection, allow a
little air space between the foam and the front of the mic grille.
Foam pop filters reduce the highs a little. So they should be left off
instrument mics, except for outdoor recording or dust protection.
Pop filters don’t reduce breathing sounds or lip noises. To get rid of
these problems, mike farther away or roll off some highs.
The most effective pop filter is a hoop with a nylon stocking
stretched over it (Figure 8.27) or a disk of perforated metal. You can
buy those, or make one with a coat hanger and a crochet hoop. Place
the filter a few inches from the mic.
Another way to get rid of pop is to put the mic at forehead height,
aiming at the mouth. This way the puffs of air shoot under the mic
and miss it. Make sure the vocalist sings straight ahead, not up at the
mic, or the mic will pop.
Wide Dynamic Range
During a song, vocalists often sing too loud or too soft. They blast
the listener or get buried in the mix; that is, many singers have a
wider dynamic range than their instrumental backup. To even out
these extreme level variations, ask the singer to use proper mic technique. Back away from the mic on loud notes; come in closer for soft
ones. Or you can ride gain on singers: gently turn them down as they
get louder, and vice versa. Usually this is done during mixdown with
automation or volume envelopes.
159
160
CHAPTER 8 Microphone Techniques
Another solution is to pass the vocal signal through a compressor,
which acts like an automatic volume control. Plug the compressor
into the vocal channel’s insert jacks or insert a compressor plug-in
in the vocal track during mixdown. A typical compressor setting for
vocals is a 2:1 ratio, ⫺10 dB threshold, and about 3 to 6 dB of gain
reduction. Of course, you should use whatever settings are needed
for the particular singer. Audio clip 26 at www.elsevierdirect.com/
companions/9780240811444 demonstrates vocal compression, as well as
breath pops and the effects of miking distance on recorded vocals.
If the singer moves toward and away from the mic while singing, the
average level will go up and down. Try to mike the singer at least
8 inches away, so that small movements of the singer won’t affect
the level.
If you must mike close to prevent leakage or feedback, ask the vocalist to sing with lips touching the foam windscreen to keep the same
distance to the mic. Turn down the excess bass using your mixer’s
low-frequency EQ (typically ⫺6 dB at 100 to 200 Hz).
Sibilance
Sibilance is the emphasis of “s” or “sh” sounds, which are strongest
around 3 to 10 kHz. They help intelligibility. In fact, many producers
like sizzly “s” sounds, which add a bright splash to the vocal reverb.
But the sibilance shouldn’t be piercing or strident.
If you want to reduce sibilance, use a mic with a flat response—
rather than one with a presence peak—or cut the highs little around
8 kHz on your mixer. Better yet, use a de-esser signal processor or
plug-in, which cuts the highs only when the singer makes sibilant
sounds. You can set a multiband compressor to act as a de-esser:
see Chapter 10 under the heading Suggested “Ballpark” Compressor
Settings.
Reflections from the Music Stand and Ceiling
Suppose that a lyric sheet or music stand is near the singer’s mic.
Some sound waves from the singer go directly into the mic. Other
sound waves reflect off the lyric sheet or music stand into the mic
(Figure 8.28, top). The delayed reflections will interfere with the
direct sound, making a colored tone quality like mild flanging.
Microphone Techniques CHAPTER 8
DIRECT
REFLECTED
NO
MUSIC
STAND
DIRECT
REFLECTED
YES
FIGURE 8.28
Preventing reflections from a music stand.
To prevent this, lower the music stand and tilt it almost vertically
(Figure 8.28, bottom). This way, the sound reflections miss the mic.
If your studio has a low ceiling, the recorded vocal might have a colored tone quality due to phase cancellations from ceiling reflections.
Try putting the mic lower and use a hoop-type pop filter. Also put
a 3-foot square of acoustic foam or pressed-fiberglass panel on the
ceiling over the singer and mic.
Vocal Effects
Some popular vocal effects are stereo reverb, echo, and doubling.
You can record real room reverb by miking the singer at a distance in
a hard-surfaced room. Slap echo provides a 1950s rock ‘n’ roll effect.
Often a vocal is mixed dry, with no reverb. A little distortion might
even be effective on some songs. You might try a vocal processor,
which offers a variety of effects. Try different EQ or different effects
on each section of a song.
Doubling a vocal gives a fuller sound than a single vocal track.
Overdub a second take of the vocal on an empty track, in sync with
the original take. During mixdown, mix the second vocal take with
the original, at a slightly lower level than the original. You can double
161
162
CHAPTER 8 Microphone Techniques
a vocal track by running it through a digital delay set at 15 to 35 msec,
or through a pitch shifter that is detuned 10 to 15 cents.
BACKGROUND VOCALS
When you overdub background vocals (harmony vocals), you can
group two or three singers in front of a mic. The farther they are
from the mic, the more distant they will sound in the recording. Pan
the singers left and right for a stereo effect. Because massed harmonies can sound bassy, roll off some lows in the background vocals.
If you want independent control of each background singer, give
each one a close-up mic and record them with separate mixer channels or separate tracks.
Barbershop, vocal jazz, or gospel quartets with a good natural blend
can be recorded with a stereo mic or stereo pair of mics about 2 to
4 feet away. If their balance is poor, close-mike each singer about
8 inches away, and balance them with your mixer. This also gives a
more “commercial” sound. If you close-mike, spread the singers at
least 2 feet apart in a circle to prevent phase cancellations.
SPOKEN WORD (SUCH AS PODCASTS AND
AUDIO BOOKS)
The tips given earlier for a lead vocalist also apply to recording the
spoken word. Be sure to keep the miking distance constant and use
a hoop-type pop filter. To prevent sound reflections into the mic, put
the script on a padded music stand that is angled almost vertically,
and put the mic in the plane of the stand near the top edge. Fold
up a corner of each script page to form a handle for turning pages
silently.
The engineer and announcer should both have the same script. Mark
the beginning of each misread sentence. The announcer should re-read
each misread sentence from the beginning to make editing easier.
CHOIR AND ORCHESTRA
Figure 8.29 shows three ways of miking a choir. If the mics will also
be used for PA, or if the venue is noisy or sounds bad, try miking
close, pan the mic as desired, and add artificial reverb (Figure 8.29A).
Microphone Techniques CHAPTER 8
MIC
2 TO 4 MICS
1.5 FT.
CHOIR
A
SIDE VIEW
FRONT VIEW
CHOIR (TOP VIEW)
B
12–20 FT.
CARDIOID
PAIR
1.5 – 3 FT.
CHOIR
CHOIR (TOP VIEW)
C
8–15 FT.
OMNI
PAIR
FIGURE 8.29
Choir miking suggestions. (A) Close-up panned mics. (B) Near-coincident stereo pair. (C) Spaced
stereo pair.
Otherwise, try a near-coincident pair of cardioid mics (Figure 8.29B)
or a pair of omni mics spaced about 2 feet apart (Figure 8.29C).
Adjust the mic-to-choir distance until you hear the desired amount
of hall acoustics in your monitors.
See Chapters 18 and 19 for suggestions on miking an orchestra.
SUMMARY
We can sum up mic placement like this: If leakage or feedback are
problems, place the mic near the loudest part of the instrument, and
add EQ to get a natural sound. Otherwise, place the mic in various
spots until you find a position that sounds good over your monitors.
There is no single “correct” mic technique for any instrument. Just
place the mic where you hear the desired tonal balance and amount
of room reverb.
Try the techniques described here as a starting point, then explore
your own ideas. Trust your ears! If you capture the power and excitement of electric guitars and drums, if you capture the beautiful timbre of acoustic instruments and vocals, you’ve made a successful
recording.
163
This page intentionally left blank
CHAPTER 9
Digital Recording
Before 1976, all music was recorded on analog tape recorders. They
record the audio signal as magnetic patterns that rise and fall in
strength as the signal waves do. Digital recorders, on the other hand,
store the audio signal as a numerical code of ones and zeros.
Let’s venture into the world of digital audio. We’ll overview how digital recording works, explore 2-track digital recorders, and explain
multitrack digital recorders. Chapter 13 covers computer recording
in detail.
DIFFERENCES BETWEEN ANALOG AND DIGITAL
Both analog and digital recorders can sound accurate—the playback
signal sounds like the input signal. But there are subtle differences. A
digital recording usually is called “clean” because it adds almost no
noise or distortion to the input signal. Some analog tape recorders
add a little “warmth” to the sound. It’s likely due to slight third harmonic distortion, head bumps (bass boost), and tape compression
at high recording levels. Compared to analog tape recordings, digital recordings have measurably less hiss, frequency-response errors,
modulation noise, distortion, unsteady pitch, and print-through.
However, those characteristics are inaudible in the best analog tape
recorders when they are properly maintained and aligned. Although
older digital recorders sounded harsh compared to analog, they
improved with each generation.
Compared to analog tape recorders, digital recorders and their media
tend to be smaller and lower cost. They can record timing information.
165
166
CHAPTER 9 Digital Recording
Also, they have random-access so you can locate a particular section of
the recording very quickly.
DIGITAL RECORDING
Like an analog tape deck, a digital recorder puts audio on a magnetic
medium, but in a different way. Here’s what happens in the most
common digital recording method—pulse code modulation or PCM:
1. The signal from your mixer, preamp, or audio interface (Figure
9.1A) is run through a lowpass filter (anti-aliasing filter),
which removes all frequencies above 20 kHz.
2. Next, the filtered signal passes through an analog-to-digital (A/D)
converter. This converter measures (samples) the voltage of the
audio waveform several thousand times a second (Figure 9.1B).
3. Each time the waveform is measured, a binary number (made
of 1’s and 0’s) is generated that represents the voltage of the
waveform at the instant it’s measured (Figure 9.1C). This process is called quantization. Each 1 and 0 is called a bit, which
stands for binary digit. The more bits that are used to represent
each measurement (the higher the bit depth), the more accurate the measurement.
4. These binary numbers are stored on the recording medium as
a modulated square wave recorded at maximum level (Figure
9.1D). For example, the numbers can be stored magnetically
on a hard disk.
The playback process is the reverse:
1. The binary numbers are read from the recording medium, such
as a hard disk (Figure 9.2A).
2. A digital-to-analog (D/A) converter translates the numbers
back into an analog signal made of voltage steps (Figure 9.2B).
3. An anti-imaging filter (lowpass filter, smoothing filter, reconstruction filter) smoothes the steps in the analog signal, resulting in the original analog signal (Figure 9.2C).
As a simplified analogy, suppose you want to reproduce the circular
shape of a wooden disk.
1. You lay a ruler across the disk every 1/4 inch and measure
how wide the disk is at each point (Figure 9.3A). That’s like
sampling.
0.6
0.6
0.4
0.4
Time
0.0
0.2
Volts
Volts
0.2
–0.2
–0.2
–0.4
–0.4
–0.6
–0.6
Sampling
interval
(B) The voltage is measured or
sampled at regular intervals.
(A) The audio waveform enters
the A/D converter.
1111
1110
1101
1100
1011
1010
1001
1000
0111
0110
0101
0100
0011
0010
0001
0000
Time
0.0
First value
1000 1011 1010 1100 1111 1100
1000 0100 0001 0011 0101 1000
Last value
(D) The numbers are stored on
the recording medium.
(C) The voltage measurements are
converted to binary numbers.
FIGURE 9.1
A/D conversion. See text. 4-bit measurements are shown for clarity, but 16- or 24-bit measurements are
standard.
1111
0.6 1110
1101
0.4 1100
Volts
0.2
First value
1000 1011 1010 1100 1111 1100
0.0
–0.2
–0.4
1000 0100 0001 0011 0101 1000
Last value
(A) The numbers are read from
the recording medium.
–0.6
1011
1010
1001
1000
0111
0110
0101
0100
0011
0010
0001
0000
(B) The binary numbers are converted
to voltage steps.
0.6
0.4
Volts
0.2
0.0
Time
–0.2
–0.4
–0.6
(C) The voltage steps are
smoothed out (filtered out).
FIGURE 9.2
D/A conversion. See text. 4-bit measurements are shown for clarity, but 16- or 24-bit measurements are
standard.
168
CHAPTER 9 Digital Recording
Ruler measurements
A
Disk
B
Strips whose
length matches
the ruler
measurements
C
Filing the strip
edges recreates
the shape of the
disk
FIGURE 9.3
An analogy of digital recording and playback: recreating a disk shape using wooden strips.
2. You cut a bunch of wood strips with the same lengths as those
measurements, and put the strips side-by-side. They look like
stair steps going up and down (Figure 9.3B).
3. You file the raggedy edges of the wood strips in an arc so they join
together smoothly. That’s like the anti-imaging filter. You’ve just
reproduced the shape of the original wooden disk (Figure 9.3C).
As we said, digital recording reduces noise, distortion, speed variations, and data errors. Because the digital playback head reads only
1’s and 0’s, it is insensitive to the magnetic medium’s noise and distortion. During recording and playback, numbers are read into a
buffer memory and read out at a constant rate, eliminating speed
variations in the rotating media. Reed-Solomon coding during
recording, and decoding during playback, corrects for missing bits by
using redundant data.
If a digital recording is on a defective medium such as a scratched
compact disc, errors (missing samples) can occur. Usually these
errors can be corrected by interpolation. This algorithm looks at
Digital Recording CHAPTER 9
data before and after the blank sample and “guesses” what its value
should be. If the errors are too extended to correct, the resulting
audio has a silent spot or a burst of noise.
Almost all digital recording devices employ the same A/D, D/A conversion process, but use different storage media: a hard-disk drive
records on magnetic hard disk, a compact disc and DVD recorder
record on an optical disc, a memory recorder records onto a Flash
memory card, and a sampler records into computer memory. The
sound quality of any of these devices depends mainly on its A/D and
D/A converters.
Digital audio is recorded on a computer hard drive as a wave file or
Audio Interchange File Format (AIFF) file. Both are standard formats
for audio files. Wave (.wav) is for PC and AIFF is for Mac. Both formats use linear PCM encoding, with no data compression (explained
in Chapter 20). Two wave formats are Riff and Broadcast wave. The
latter facilitates interchange of program material between audio
workstations.
Bit Depth
As we said, the audio signal is measured many thousand times a second to generate a string of binary numbers (called words). The longer each word is (the more bits it has), the greater the accuracy of
each measurement. Short words give poor resolution of the signal
voltage (high distortion); long words give good resolution (low distortion). Bit depth or resolution are other terms for word length.
In our wooden disk analogy, bit depth (word length) corresponds to
the accuracy of the ruler measurement: whether we measured to the
closest 1/8”, 1/16”, 1/32”, etc.
A word length of 16 bits is adequate (but not optimum) for hi-fi
reproduction. It is the current standard for CDs. Most digital recorders offer 24-bit word lengths. More bits sound smoother and more
transparent, but need more disk storage space.
Even though CDs are a 16-bit format, they sound better when
made from 24-bit recordings. During mastering, you add dither
(low-level noise) to the 24-bit recording, then export or save it as a
16-bit recording. You copy the 16-bit recording to CD. The dither
helps the 16-bit recording sound more like the 24-bit recording.
169
170
CHAPTER 9 Digital Recording
In other words, dither lets you retain most of the quality and resolution that you recorded at 24 bits, even though the recording ends up
on a 16-bit CD. We explain dither later in this chapter.
Suppose you record with a 24-bit word length. All signal levels in
that recording have 24-bit resolution. Even low-level signals use all
the bits—but most of the bits are zeroes.
Sampling Rate
Sampling rate or sampling frequency is the rate at which the A/D
converter samples or measures the analog signal while recording.
For example, a rate of 48 kHz is 48,000 samples per second; that is,
48,000 measurements are generated for each second of sound. The
higher the sampling rate, the wider the frequency response of the
recording. According to the Nyquist-Shannon theorem, the upper
frequency limit of a digital recording is one-half the sampling rate.
Compact discs use a 44.1 kHz sampling rate, so their frequency
response extends to 22.05 kHz.
In our wooden disk analogy, sampling rate corresponds to the number of wood strips you cut to form the disk. If you measure the disk
every 1/2” across, that is like using a low sampling rate. If you measure the disk every 1/8” across, that is like using a high sampling rate.
Sampling rate for high-quality audio can be 44.1, 48, 88.2, 96, or
192 kHz. CD quality is 44.1 kHz/16 bits. A 96 kHz sampling rate
can be used on the DVD. State-of-the-art is Super Audio CD or linear PCM at 192 kHz/24 bits (but you’re more likely to see 96 kHz/24
bits).
According to some sources, higher sampling rates sound smoother
and more transparent but need more disk storage space and a faster
hard drive. One study in the September 2007 issue of the Journal
of the Audio Engineering Society has shown that rates higher than
44.1 kHz yield no audible improvement over 44.1 kHz.
Digital recording is based on the Nyquist-Shannon theorem. It
defines the “Nyquist frequency” as half the sampling rate. If the sampling rate is 44.1 kHz, the Nyquist frequency is 22.05 kHz.
The theorem proves that the original analog signal can be reconstructed exactly from its samples—if the highest frequency in the
Digital Recording CHAPTER 9
signal is less than the Nyquist frequency. That happens when the
original signal is sent through a lowpass (high-cut) filter to remove
frequencies above the Nyquist frequency.
If the analog signal is not filtered, aliasing occurs: high frequencies
above the Nyquist frequency appear as lower frequencies. In other
words, inaudible ultrasonic frequencies are converted to audible
tones. That’s why an anti-aliasing filter is needed.
If you lowpass filter the analog signal to the Nyquist frequency
before it’s sampled, there is only one waveform that can pass
through the sample points, and that is the original analog waveform.
Figure 9.1B shows the sample points and the waveform that passes
through them.
Current A/D and D/A converters use a process called oversampling
to improve the sound. Oversampling is sampling an audio signal
at a higher rate than needed to reproduce the highest frequency in
the signal. For example, sampling a 20 kHz audio signal at 8 times
44.1 kHz is called “8x oversampling.” This process is followed by a
digital lowpass filter and a gentle-slope analog anti-alias filter. The
result is less phase shift and less harshness compared to a steep,
“brick-wall” analog filter used alone.
In summary, a digital audio system samples the analog signal several
thousand times a second, and quantizes (assigns a value to) each
sample. Sampling rate affects the high-frequency response. Bit depth
affects the dynamic range, noise, and distortion.
An alternative to PCM is 1-bit (bitstream) encoding with a
2.8224 MHz or higher sampling rate. This is the Direct Stream
Digital (DSD) process used in the Super Audio CD and in some Korg
portable digital recorders. DSD offers a frequency response from DC
to 100 kHz with 120-dB dynamic range and a very smooth, analoglike sound.
Most recent digital devices use delta-sigma A/D converters, which
produce 1-bit streams at a very high sample rate (typically using
64x oversampling). Then a digital lowpass filter removes the aliasing and quantizing noise. Finally, a decimation circuit downsamples
(converts) the bitstream to 16- or 24-bit PCM samples at the desired
sample rate (44.1 kHz, 96 kHz, and so on).
171
172
CHAPTER 9 Digital Recording
With delta-sigma A/D conversion, recorded sample rates above
44.1 kHz don’t improve the sound quality because the analog input
signal is 64x oversampled, no matter what the recorded sample rate is.
In a digital transmission, the two channels of a stereo program are
multiplexed. That is, one word from channel 1 is followed by one
word from channel 2, which is followed by one word from channel 1,
and so on.
Data Rate and Storage Requirements
The data rate of digital audio (in bytes per second) is
Bit depth/8 ⫻ Sampling rate ⫻ Number of tracks.
Divide by 1,048,576 to get megabytes (MB) per second. For example,
the data rate of a 24-bit/44.1 kHz recording of 16 mono tracks is
(24 / 8 ⫻ 44,100 ⫻ 16) /1, 048, 576 ⫽ 2 MB/sec.
Recording digital audio on a hard drive consumes a lot of space. The
storage required is
Bit depth/8 ⫻ Sampling rate ⫻ Number of tracks
⫻ 60 ⫻ Number of minutes
Divide by 1,048,576 to get MB. Divide MB by 1024 to get gigabytes
(GB). For example, suppose you record a concert at 24 bits, 44.1 kHz,
16 tracks, for 2 hours. The hard-drive space needed is
(24 /8 ⫻ 44,100 ⫻ 16 ⫻ 60 ⫻ 120)/1, 048, 576
⫽ 14, 534.9 MB or 14.2 GB
Note that the operating system might report the capacity of the disk
drive up to 7% lower than it really is.
Table 13.1 shows the amount of storage needed for a 1-hour recording. Not shown in Table 13.1, a 1-bit stereo recording at 2.8 MHz
consumes about 42.3 MB/minute or 2.3 GB per hour.
Digital Recording Level
In a digital recorder, the record-level meter is a peak-reading LED or
LCD bar graph meter (Figure 9.4) that reads up to 0 dBFS (FS means
full scale). In a 16-bit digital recorder, 0 dBFS means all 16 bits
Digital Recording CHAPTER 9
Levels over 0 dBFS
cause clipping distortion
Recommended
maximum
peak recording level
OVER
0 dBFS (all bits on)
–6 dB
–12 dB
–18 dB
The height of this bar
shows the recording
level
–24 dB
–30 dB
–36 dB
–42 dB
FIGURE 9.4
Digital record-level meter.
are on (1) at the waveform peak, and all 16 bits are off (0) at the
waveform trough. In a 24-bit digital recorder, 0 dBFS means that all
24 bits are on or off.
The OVER indication means that the input level exceeded the voltage
needed to produce 0 dBFS, and there is some short-duration clipping
of the output analog waveform. This clipping can sound really nasty.
Some manufacturers calibrate their meters so that 0 dBFS is less than
16 or 24 bits on; this allows a little headroom. When you set the recording level, it’s a good idea to aim for ⫺6 dB maximum so that unexpected
peaks don’t exceed 0 dBFS. If you’re making a 24-bit recording, the
recording level isn’t very critical because a 16-bit signal is at ⫺48 dBFS!
THE CLOCK
Each digital audio device has a clock or internal oscillator that sets the
timing of its samples. The clock signal is a series of pulses running at
173
174
CHAPTER 9 Digital Recording
the sampling rate. When you transfer digital audio from one device to
another, their clocks must be synchronized. One device must provide
the master clock and the other must be the slave. This clock comes
embedded in the digital signal, or comes on a separate wire or connector as a word clock signal.
In your digital devices or DAW, look for a switch or software menu
that selects the clock source. The three choices are internal clock,
external clock, or word clock. Internal clock is the A/D converter’s
clock. External clock (also called digital input) is the clock signal
embedded in another device’s digital audio signal. Word clock is a
separate cable with timing signals. Set the master device to internal
clock and set the slave devices to external clock (digital input). If
you’re using a master word-clock generator feeding the word-clock
connectors of other devices, set the slave devices to word clock.
Some receiving devices synchronize automatically to the embedded
clock from the source device. In that case you don’t need to set the
clock source.
Your system’s A/D converter or interface is usually the best master
clock. Connect its word clock output (if any) to the input of a wordclock splitter or distribution unit. Connect a cable from each output
of the distribution unit to the word-clock input of each of the other
devices. That way all the devices will be synchronized.
Word-clock cables under 25 feet should be unbalanced 75-ohm
cables with BNC connectors. Word-clock cables over 25 feet should
be balanced 110-ohm AES digital cables with XLR connectors,
although regular mic cables can work.
In large recording systems where you mix or assemble digital audio
signals from a variety of sources, it’s convenient to use a separate
master clock such as Apogee’s Big Ben, Mutec’s iClock, or Drawmer’s
M-Clock. Most master clocks have several word-clock outputs. Connect each one to each slave device in your studio to keep them in
sync. The receiving device’s word-clock connector must be terminated
properly—refer to the user’s manual. If you daisy-chain the wordclock connectors between audio devices (not recommended), only
the last device in the chain should be terminated.
Digital Recording CHAPTER 9
DIGITAL AUDIO SIGNAL FORMATS
There are seven major formats for carrying digital-audio signals
between various devices: AES3, MADI, S/PDIF, ADAT Lightpipe,
TDIF, USB, and FireWire. With all these formats you connect a single
cable between digital out and digital in. Let’s look at each format.
■
■
■
■
■
■
AES3 (also called AES/EBU or IEC-60958 Type I): This professional format transfers 2 channels of digital audio over a
110-ohm shielded twisted-pair cable with XLR connectors. That
is not the same as a mic cable. An AES3 cable can be up to
100 meters long. AES3id uses 75 ohm coaxial cable and BNC
connectors for runs up to 1000 meters. You need one AES3
interface for each pair of channels.
MADI (Multichannel Audio Digital Interface or AES10-2003):
This professional format transports 28, 56, or 64 channels up
to 24 bits/96 kHz on 75-ohm coaxial cable or fiber-optic cable.
It’s often used for linking large mixing consoles to digital multitrack recorders. The cable may be up to 3000 meters long.
S/PDIF (Sony/Philips Digital Interface; also called IRC-0958
Type II): This is a 2-channel consumer or semi-pro format. It uses
a single 75-ohm coaxial cable up to 10 meters long with RCA or
BNC connectors, or a fiber-optic cable with TOSLINK connectors. Optical interfaces prevent ground loops and cable losses.
ADAT Lightpipe: The Alesis ADAT digital multitracks use a
Lightpipe, which sends 8 channels of digital audio on a single optical cable with TOSLINK connectors. Every 8 channels
of transfer require a separate Lightpipe cable. Data transfer is
up to 24 bit/48 kHz or 24 bit/96 kHz with half the number of
channels.
TASCAM TDIF (TASCAM digital interface): The TASCAM DA-88
and similar modular digital multitracks use a multiconductor
cable with standard DB-25 connectors. TDIF sends 8 channels
of digital audio in and out on a single cable, which can be up
to 5 m long.
USB (Universal Serial Bus) and FireWire (IEEE 1394) are standard Mac/PC protocols for high-speed serial data transfer
between a computer and an external device, such as a hard drive,
USB drive, MIDI interface, or audio interface. More details are in
Chapter 13 under the heading FireWire or USB Audio Interface.
175
176
CHAPTER 9 Digital Recording
AES3, S/PDIF, and ADAT Lightpipe are self-clocking systems: the clock
is embedded in the signal and marks the start time of each sample.
MADI, TDIF, and ADAT Sync carry a separate word-clock signal on a
separate connector or wire. Note that FireWire and USB signals don’t
have clock information; they are reclocked at the receiving device.
Converting Signal Formats
AES and S/PDIF signals are similar but not necessarily compatible.
You can convert one to the other using a format converter. Some
sound cards and digital mixers do this conversion. Lightpipe and
TDIF signals can be converted as well.
Some digital audio devices don’t implement AES or S/PDIF correctly,
so they don’t interface with some other devices.
DITHER
Your HD recorder, software, and digital mixer may run at 24 bits, but
the result ends up on a 16-bit CD. When you save a 24-bit audio
file as a 16-bit file to transfer to CD, those last 8 bits are truncated
or cut off. The result may be a grainy static sound at very low levels.
This distortion can be prevented by adding low-level random noise
(dither) to the signal. Let’s explain.
A 24-bit resolution can accurately capture the quietest parts of a musical program: very low-level signals such as the end of long fades and
reverb tails. But truncation of that signal to 16 bits makes those lowlevel signals sound grainy or fuzzy, because 16 bits is a less accurate
measurement of the analog waveform than 24 bits. This fuzzy sound,
called quantization distortion, doesn’t exist at normal high levels.
What causes this distortion? Each digital word is made of a certain
number of bits. During quantization, the A/D converter assigns the
closest possible digital number to represent the measured voltage of
each sample. The last or rightmost bit (least significant bit or LSB)
switches on or off depending on whether the converter rounds the
word value up or down. If this switching occurs in the 16th bit, it
may be faintly audible as a fuzzy noise during quiet passages.
Also, a 24-bit recording has 256 possible levels in the lower 8 bits.
But after the signal is truncated to 16 bits, that resolution is lost.
Digital Recording CHAPTER 9
To solve this problem, dithering adds random noise (random 1’s and
0’s) to the lowest 8 bits of the 24-bit signal (at about ⫺100 dB) before
they are truncated to 16 bits. That noise modulates the 16th bit with
some 24-bit information (bits 17 through 24) in the form of pulsedensity modulation. The average value of that modulated square wave
is recovered by a lowpass filter. Then most of the 24-bit sound quality
is restored, and the quantization distortion changes to a smooth hiss.
To make that added hiss less obvious, noise shaping is used. Noise
shaping applies an oversampling filter to the noise, which reduces its
level in the midrange where our ears are most sensitive, and increases
its level in the high frequencies where it’s less audible.
Compared to a truncated signal, a truncated-and-dithered signal sounds slightly more transparent. Fades and reverb tails sound
smoother, and there’s more sonic detail. Signals below the noise
floor become audible.
For best sound quality, apply dither only once when you convert a high
bit-depth source to its 16-bit CD format. For example, record at 24 bits
and stay there through the entire project. Don’t re-dither material that
has already been dithered—switch off any dithering. Add dither as the
very last step when you export your stereo mix to 16 bits.
To hear the effects of dithering, start with a clean 24-bit recording,
reduce its level 50 dB in your editing software, and export it as a
16-bit file. Export it in three ways: without dither, with dither, and
with noise shaping added. Next, normalize the exported recordings
so that the highest peak hits 0 dBFS. Then listen to the resulting 16-bit
files at high level over headphones. Compare the processing and use
whatever sounds best. Generally, a signal that is truncated with no
dither is accompanied by a rough, grainy noise. A signal that is truncated and dithered is accompanied by smooth, quiet hiss or silence.
With good dithering algorithms, it’s possible to preserve most
of the 24-bit quality (ultra-low distortion, fine detail, and ambience) when converting to 16 bits. One such system is the POW-rTM
Psychoacoustically Optimized Word-Length Reduction algorithm from
the Pow-r Consortium LLC. It reduces the high-resolution, higher
word lengths (20 to 32 bit) to a CD-standard, 16-bit format while
retaining the transparency of the high-resolution recording. In other
words, the 16-bit CD sounds like the original 24-bit recording.
177
178
CHAPTER 9 Digital Recording
JITTER
Accurate A/D and D/A conversions rely on the clock precisely sampling the analog signal at equal time intervals. Jitter is the unstable timing of samples that occurs in A/D and D/A conversion. Any
change between the sample times, even nanoseconds, creates amplitude errors—small changes in the audio waveform’s shape—resulting in a slight veiling of the sound (low-level distortion or noise).
Some researchers claim that a jitter spec under 250 nanoseconds is
inaudible; others say 20 nanoseconds; others say 250 picoseconds.
During A/D and D/A conversion, one cause of jitter is analog noise
and crosstalk in the recording system. They affect the switching times
and switching threshold of the clock, causing frequency modulation
of the clock. They also affect the analog filters and oscillators used in
the clock’s phase-locked loops (PLLs).
Jitter is also caused by inadequate cables. They pick up hum and
noise, and introduce phase shift and high-frequency attenuation,
which degrade the timing of the digital signal. However, the AES3
and ADAT interfaces can handle a lot of timing variation, so cableinduced jitter is almost never a problem.
In theory, jitter doesn’t occur during real-time data transmission
over FireWire or USB because they are clockless systems. In other
words, FireWire and USB connections don’t introduce jitter on their
own. FireWire data transfers are “isochronous,” which means these
data must be delivered at a certain minimum data rate. Digital audio
and video signal streams require isochronous data flow to ensure
that audio data are delivered as fast as they are played and recorded.
However, there’s a potential problem. Part of the FireWire and USB
signals are sync-message packets. FireWire and USB interfaces can
slightly vary the timing of those messages. Then, when the receiving
device extracts the clock with a PLL, jitter can be created.
To reduce jitter:
■
■
Use high-quality clock sources with low jitter specs (under
1 nanosecond). Some audio interfaces and A/D converters are
specially designed to reduce jitter, and usually include a PLL.
Use high-quality, well-shielded cables designed for digital signals, and as short as possible. Don’t use mic cables for AES3;
don’t use hi-fi cables for S/PDIF.
Digital Recording CHAPTER 9
■
■
Keep analog and digital cabling separate.
Use your audio interface or A/D converter as the master clock.
Don’t drive it from an external clock source.
DIGITAL TRANSFERS AND COPIES
When you send a digital audio signal in real time from one device
to another, they must be set to the same sampling rate. For example,
if the digital signal from a multitrack recorder is 48 kHz, the digital
mixer it feeds must also be set to 48 kHz. In a digital transfer, the
sending device usually is the master, and the receiving device is the
slave. Normally you can set the slave to automatically lock onto
the sampling rate of the master. Set the master device to Internal
Clock, and set the slave device to External Clock. Or set the slave’s
clock source to whatever device the master is.
The word length (number of bits) of digital transfers is less critical.
As long as the two machines handle the same digital format (AES,
S/PDIF, etc.), transfers of any word length can be done.
It’s okay to send a lower bit signal to a higher bit device; for instance,
feed a 16-bit signal from a DAT into a 24-bit digital mixer. The mixer
will add more zeroes to fill out the digital word, with no effect on
sound quality. It’s also no problem to feed a device more bits than
it can hold. The receiving device ignores the extra bits. For example,
if you send a 24-bit signal to a 16-bit recorder, the last 8 bits will be
truncated or cut off.
Truncation adds slight distortion, but this can be reduced by adding
dither—low-level noise—to the signal before truncating. So it’s best
to keep the word length as long as possible as you’re working on a
project. Always apply dither before you reduce the word length, and
only then.
What if you have a 48 K recording that you want to release on CD?
The CD mastering engineer can (1) convert the 48 K digital signal to
44.1 K, which might degrade the sound, or (2) use the analog output
of his playback machine, and convert the analog signal to digital at
44.1 K. This may or may not sound better.
Digital audio transfers to or from DAT or MDM tape, and to or from
CD-Rs, may or may not create perfect clones because of inadequate
error correction or interpolation errors.
179
180
CHAPTER 9 Digital Recording
When you copy a WAVE or AIFF file rather than a digital audio signal, the copy is a perfect clone of the original file. The file transfers
much faster than the real-time playback of the signal. Flawless file
copies can be made in these ways:
■
■
■
■
Inside a computer from one hard drive location to another
From one computer’s hard drive to another computer’s hard
drive
Via Ethernet, USB, FireWire, or the Internet
Between computers via data CD-R, Jaz, or Zip drive; floppy
disks; or Flash memory cards.
The following data compression algorithms don’t lose any data, so
they don’t degrade audio quality:
■
■
■
■
■
■
Emagic’s Zap
Waves’ TrackPac
PKWare’s PKZIP
WinZip Computing’s WinZip
Aladdin Systems’ StuffIt
Meridian Lossless Packing (MLP).
2-TRACK DIGITAL RECORDERS
Now that digital recording has been overviewed, let’s look at digital
recorders. They come in 2-track and multitrack formats. The 2-track
formats are
■
■
■
■
Portable hard-drive recorder
Digital Audio Workstation; this is a computer (usually a laptop) with a sound card or other audio interface, plus recording
software
CD-R recorder
Flash memory recorder.
We’ll look at each one. No matter which format you use, the sound
quality depends on the A/D converter and mic preamp (if any).
Portable Hard-Drive Recorder
This device records onto a built-in hard drive. For example, the
Edirol R4 records up to 4 tracks simultaneously onto a 40 GB drive,
Digital Recording CHAPTER 9
either MP3 or WAV files up to 24 bit/96 kHz. The unit has XLR/phone
inputs with phantom power, built-in stereo speakers, USB 2.0 interface to a computer, waveform editing, limiting, and a compact-flash
slot. See CD www.edirol.com.
The Digital Audio Workstation
The Digital Audio Workstation (DAW) is a computer running recording
software with a connected audio interface such as a sound card (Figure
9.5). A DAW allows you to record, edit, and mix audio programs
entirely in digital form. It can store up to several hours of digital audio
or MIDI data. You can edit this data with great precision on a computer
monitor screen. What’s more, you can add digital effects and perform
automated mixdowns. We’ll cover DAWs in detail in Chapter 13.
A laptop computer can make a great portable 2-track digital recorder.
Just add some recording software and an audio interface connected
by USB or FireWire.
MONITOR
HARD
DRIVE
PCI SLOT, USB PORT,
OR FIREWIRE PORT
COMPUTER
KEYBOARD
ALESIS
FIREPORT
AUDIO INTERFACE
SOUND CARD
AUDIO INTERFACES (CHOOSE ONE)
FIGURE 9.5
A computer Digital Audio Workstation (DAW).
MOUSE
CONTROLLER OR MIXER
WITH I/O
181
182
CHAPTER 9 Digital Recording
CD Recordable (CD-R)
Another form of digital recording is the compact disc. On your own
desktop, you can cut a CD by using a CD-R recorder (CD burner). It’s
exciting to hear one of these CDs playing your music with the purity
of digital sound. The sound quality meets or exceeds CD standards.
CD-R stands for compact disc recordable. This optical medium
is a write-once (nonerasable) format. CD-RW stands for compact disc re-writable; you can erase it and record a new program.
Although any modern CD player or CD-ROM drive will play a CD-R,
many CD players—especially older ones—can’t play CD-RW discs.
Most new CD-ROM drives support CD-RW, but not all will read
CD-RW discs at full speed. CD-RW blank discs cost more than CD-R
blank discs.
How can you use the CD-R format? You could make demos for your
own band, or make a one-off copy of your stereo mixes for clients.
Use the CD-R as a pre-master to send to a CD replicator. Another
function is to compile sound libraries of production music, samples,
and sound effects. In addition to recording audio, CD-Rs can store
data files such as WAVE, AIFF, or MP3 files to send to other studios.
If handled and stored with care, the CD format can archive your
recordings (but always use another medium as a backup).
CD-R FORMATS
First you’ll be faced with two basic choices of CD-R recorder:
1. A stand-alone CD-R recorder, sometimes called a consumer
CD recorder
2. A computer peripheral CD-R recorder, also called a computer
CD burner; you plug it into your computer system, or buy a
computer with a CD burner built in.
Both types produce discs that sound equally good. Both types of
discs will play on any audio CD player.
The stand-alone CD recorder (Figure 9.6) has everything you need
in one chassis. Inside is a CD transport, laser, and microprocessor.
On the back are analog and digital ins and outs. On the front are
the level meters, record-level knob, display, and keypad. Because the
Digital Recording CHAPTER 9
FIGURE 9.6
TASCAM CD-RW901SL, an example of a stand-alone CD recorder.
stand-alone unit needs no external computer, it’s user-friendly. Just
connect your audio source such as a mixer output or an analog or
digital recording. Set the recording level and start recording.
The stand-alone unit can write audio but not data. The 63-minute
blank discs it uses are the “Music CD-R” or “Digital Audio CD-R”
format. Prices for stand-alone CD-R writers start around $250.
A computer CD recorder costs less: about $35 and up. The unit plugs
into an EIDE or SCSI connector in your computer. It can write audio
or computer data. Its blank discs, called “CD-R” or “Data CD-R,”
cost less than 35 cents in quantity. (Music CD-Rs will work in a computer CD burner as well.) Disc length is 74 minutes (650 MB) or 80
minutes (700 MB). Some discs permit recording at up to 52 times
real time if your computer and hard drive are fast enough.
The computer CD-R recorder also requires a CD recording program,
usually packaged with the recorder. You don’t have to use that program; other ones are available that you might prefer. Be sure the program is compatible with your recorder. You’ll also need a sound card
and some software to record audio onto hard disk.
Multi-drive CD burners let you record or copy several CDs at once.
CD-R TECHNOLOGY
While conventional CD players follow the Sony-Philips Red Book
standard, CD-Rs conform to the Orange Book part II standard. Once
recorded, a CD-R disc meets the Red Book standard.
183
184
CHAPTER 9 Digital Recording
A recordable CD is the same size as a standard CD, but it is more
colorful. On top is a metal reflective layer; on the bottom is a recording layer made of blue cyanine dye or yellow (gold) phthalocyanine
dye. The blue layer appears green because of the gold layer behind
it. Yellow dye lasts a little longer in accelerated aging tests, and it
may work better with high-speed CD-R drives. A few other colors are
available as well.
A blank CD-R is made of four layers (Figure 9.7):
1. Clear plastic (protects the metal layer)
2. Metal layer (gold, silver, or silver alloy, which reflects the laser
light)
3. Dye (for the recording)
4. Clear plastic (protects the dye layer).
The dye fills a spiral groove that is etched in the bottom clear-plastic
layer. This groove guides the laser.
To record data on a disc, the laser melts holes in the dye layer. The
plastic layer flows into the holes to form pits. During playback,
Top clear
plastic layer
CD SIDE VIEW
Dye layer
FIGURE 9.7
CD-R construction.
Laser beam reflects off
metal layer through pits
in dye layer
Metal layer
Bottom clear
plastic layer
Digital Recording CHAPTER 9
the same laser reads the disc at lower power. At each pit, laser light
reflects off the metal layer. The reflected light enters the laser reader,
which detects the varying reflectance as the pits go by.
In contrast with a standard CD, a CD-R disc has two more data
areas:
■
■
The program calibration area (PCA). The CD recorder uses
this area to make a test recording, which determines the right
amount of laser power to burn the disc (4 to 8 milliwatts).
The program memory area (PMA). This area stores a temporary
table of contents (TOC) as the CD-R tracks are assembled. The
TOC is a list of the tracks, their start times, and the total program time. The recorder uses the PMA for this information until
it writes the final TOC.
According to CD-R manufacturers, the expected lifetime of a CD-R is
about 70 to 100 years with careful handling. Avoid sunlight and high
temperatures. Be sure not to damage the label side of the disc by
writing on it with a ballpoint pen or pencil—use a soft felt-tip pen
that is water-based, such as a CD-marking Sharpie. Store CDs vertically and keep them in their cases. Avoid labels because the adhesive
can attack the protective plastic layer, and the paper might warp over
time. CD-RW life expectancy is claimed to be about 25 years.
CD-R SESSIONS, DISC-AT-ONCE, AND TRACK-AT-ONCE
A session on disc is made up of a lead-in, program area, and leadout. Each session has its own TOC. Each lead-in and lead-out consumes 13 MB of disc space.
With the multisession feature, you can write several sessions on a
disc at different times. This feature comes in handy when you need
to add information to a disc a little at a time. Only the first session
on the disc will play on an audio CD player, so the discs are just for
your own use and not for distribution. On a multisession disc, you
lose about 23 MB of space when you close the first session and about
14 MB for each following session.
CD-R recorders also permit Disc-At-Once recording, in which the
entire disc must be recorded nonstop. You can’t add new material
once you write to the disc. With some software, Disc-At-Once lets you
185
186
CHAPTER 9 Digital Recording
set the length of silence between tracks (down to 0 seconds), and lets
you control how the tracks are laid out on disc. Use Disc-At-Once
for all pro audio work.
Most CD-R recorders allow Track-At-Once recording. They can record
one track (or a few tracks) at a time—up to 99 tracks. You can play a
partly recorded disc on a CD-R recorder, but the disc won’t play on a
regular CD player until the final TOC is written. Track-At-Once isn’t
recommended for audio because it puts 2-second spaces and clicks
between audio tracks.
If you want no pauses between tracks (as on a live album), get a CD-R
writer with Disc-At-Once. Also get some software that can adjust the
pause length down to zero, or that can set the start ID of each song
anywhere in the program. Note that a stand-alone CD-R recorder will
copy your edited program as it is, with or without pauses.
A good CD-R burner has a buffer of at least 2 MB. Some burners include SCMS copy code. Be sure that you can return the CD-R
burner if it proves to be unreliable.
HOW TO USE A STAND-ALONE CD-R WRITER
Let’s say that you have an edited recording of song mixes on a 2-track
recorder, and you want to copy them onto a CD by using a selfcontained CD-R writer.
Connect the 2-track recorder’s output to the CD-R writer’s input.
Use a Music CD-R or Digital Audio CD-R. Set your levels and begin
recording. Your program will copy to disc in real time.
Depending on the CD-R writer, the recording’s start IDs may or may
not convert to CD track numbers. If not, you can use a converter box,
or add the track numbers manually while you record.
Using a computer CD-R burner is covered in Chapter 15. For more
information on CD-R technology, see www.cdrfaq.org and www
.harmony-central.com/Features/CDRecorder/.
Flash Memory Recorder
This is a portable 2-track digital recorder that records onto a Flash
memory card such as a Compact Flash or Secure Digital (SD) card.
It records MP3 or uncompressed PCM WAV files (which are CD quality
Digital Recording CHAPTER 9
or better). You can copy the recorded files to a computer hard drive
through a USB cable.
Available mic connectors are XLR, 1/4” phone or 1/8” phone, with
or without 48V phantom power or plug-in power for the mics.
Options are sample rate (44.1 to 96 kHz), battery life (4 to 19
hours), battery type (alkaline, lithium, or NiMH), battery rechargeable or standard, memory size (up to 4 GB), USB version (1 or 2),
mics (built-in or not), and extra features such as a guitar tuner, vocal
cancellation, pitch change, overdubbing, CD burner, and real-time
audio interfacing to a computer. The Korg MR-1 and MR-1000 record
in 1-bit DSD format rather than PCM.
Examples of models are in Chapter 4 under the heading Portable
2-Track Recorder, and one is shown in Figure 9.8.
FIGURE 9.8
Zoom H2 Flash memory recorder.
187
188
CHAPTER 9 Digital Recording
MULTITRACK DIGITAL RECORDERS
So far we’ve covered 2-track digital recorders. Now we’ll consider the
multitrack digital recorders listed below:
■
■
■
■
Computer DAW (covered earlier)
Stand-alone hard-disk recorder
Recorder–mixer
DVD recorder (see Chapter 19).
Hard-Disk (HD) Recorder
This digital multitrack recorder was described in Chapter 4 under the
heading Separate Multitrack Recorder and Mixer.
Recorder–Mixer (Digital Personal Studio)
This device was explained in Chapter 4 under the heading Digital
Multitracker (Recorder–Mixer). An example is shown in Figure 9.9.
Listed below are some features to look for.
FIGURE 9.9
TASCAM DP-02, an example of a digital personal studio.
Digital Recording CHAPTER 9
■
■
■
■
■
■
■
■
■
■
■
■
■
■
■
Type of analog inputs and outputs. Balanced XLR or TRS phonejack mic connectors are preferred over unbalanced (TS) phonejack connectors. Balanced connections let you run longer mic
cables without picking up hum.
Type of digital inputs and outputs. S/PDIF is a common format.
Number of mic inputs: 2 and up. If you’re recording an entire
band or drum set at once, you’ll need at least 8 mic inputs.
Number of mixer channels: 8 and up.
Number of tracks: 4 to 36. The more tracks you have, the more
instruments and vocals you can record on individual tracks.
Number of tracks that can be recorded simultaneously: 8 to 24.
If you want to record an entire band at once, you need to record
many tracks at the same time.
A/D converter’s bit depth: 16 to 24 bit. The more bits, the
higher the sound quality.
A/D converter’s sampling rate: 44.1 to 96 kHz. The higher the
sampling rate, the higher the sound quality, but 44.1 kHz is
often good enough for projects that will end up on a CD. The
latest units record with 24-bit resolution and up to 96 kHz sampling rate. Some record 24 tracks of 24-bit programs at sample rates of 44.1 and 48 kHz. They record 12 tracks of 24-bit
recording at 88.2 and 96 kHz.
Number of built-in digital effects: Up to 8 stereo or 16 mono.
Examples of effects are reverberation, echo, flanging, chorus,
and compression. They add sonic excitement and a professional touch to your productions. Some recorder–mixers have
an expansion card that accepts downloadable plug-in effects.
Types of EQ: 2- or 3-band, fixed or parametric. EQ means tone
control. A 2-band EQ adjusts bass and treble, and a 3-band
adjusts bass, midrange, and treble. A fixed-frequency EQ is less
flexible than a parametric or sweepable EQ, which lets you
adjust the frequency range you want to work on.
Phantom power for condenser mics.
Backlit LCD display (bigger is better) with a waveform display.
Built-in CD burner.
Backup or transfer recordings to computer via USB.
Data compression: Recordings can be made in MP3 format as
well as WAV format. The WAV format is preferred for higher
sound quality.
189
190
CHAPTER 9 Digital Recording
■
■
■
■
■
■
■
■
■
■
■
■
■
Footswitch so you can record over mistakes (punch-in) with
your foot while playing your instrument.
Pitch control. Adjust the recording speed up or down so you
can match the pitch of recorded tracks to the pitch of new
instruments to be recorded.
Backup options: Removable hard drive or CD-R.
Levels of undo: Up to 1000. The Undo function lets you undo
an editing change you just did. If you don’t like how an edit
sounded, you can press Undo and go back to the way things
were before you made the change.
Number of locate points: Up to 1000. A locate point is a point
in time in the recording that you can have the recorder memorize and locate later. For example, you might want to mark the
time where the beginning of a song is, so you can tap a button
and go there instantly.
Expansion ports for SMPTE and MTC sync, MDM interface, and
extra ins and outs. Beginners probably don’t need these features,
but they let you connect your multitracker to other devices.
MIDI implementation: MIDI Machine Code (MMC), Control
Change (CC). These features let you control the mixer via MIDI
commands.
Types of synchronization: SMPTE, Word Clock, MTC, MIDI
Clock with Song Position Pointer, ADAT, DA-88, and RS-422.
These are various formats that you might want your multitracker to sync with.
Jog/shuttle wheel to “scrub” audio–play it slowly forward and
backward to locate an edit point.
Remote control.
Automated mixing. A hard-drive or memory recording in
the multitracker remembers your song-mix settings and mix
changes. The next time you play the song, the mixer resets to
those settings automatically. Some multitrackers have motorized faders, so that the faders move up and down just as they
did when you adjusted them while mixing.
Editing (cut and paste, etc.). This lets you do all sorts of things
with the song arrangement. You might remove parts of songs
that you don’t want to keep, loop a drum part so that it repeats
continuously, or copy a chorus to several points in a song.
Video output that plugs into a computer monitor screen.
During editing, it’s a lot easier to see details in the sound waveform if you can plug a large monitor into your multitracker.
Digital Recording CHAPTER 9
■
■
Mastering tools and CD-burning ability. In mastering you
compile a playlist of song mixes from which you can create an
album. Some multitrackers include programs that let you record
your songs on a CD-R recorder, either external or built-in.
Number of virtual tracks (recordings of separate takes of the
same instrument): 8 and up.
A virtual track is a 1-channel recording of a single take or performance on a random-access medium, such as a hard drive. You can
record several virtual tracks or takes of a single instrument, then
select which take you want to hear during mixdown. For example,
suppose you have a 16-track hard-disk recorder. You might record 8
or more virtual tracks (takes) of a vocal on hard disk. Then during
mixdown, choose which virtual track will play as one of those 16
tracks. Some recorders can be set up to select parts of different virtual
tracks during playback—a process called comping. For example, play
vocal take 15 during the verse, play vocal take 3 during the chorus,
and so on.
In other words, you can associate several virtual tracks, or takes, with
a single “real” track. You might play 8 tracks at once, and you can
choose which virtual track (take) plays on each track. For example,
you can record several takes of a guitar solo—keeping each one—
and choose the best take during mixdown. You also can create a
composite track, which is made of the best parts of several takes. All
HD recorder–mixers let you record virtual tracks.
Pros and Cons of Three Multitrack Recording Systems
Let’s compare a hard-drive recorder, recorder–mixer, and computer DAW.
Multitrack Hard-Drive Recorder and Separate Mixer
Pros:
■
■
■
■
■
■
The recorder is portable—great for on-location recording.
Real faders and knobs.
Can export audio data to a computer for editing, or can use its
own editing firmware.
Stable. Less likely to crash than a computer recording system.
This is a big advantage both in on-location recording and studio recording.
The mixer allows easy set up of monitoring and cue mixes.
The recorder might have higher quality converters than a computer recording system.
191
192
CHAPTER 9 Digital Recording
Cons:
■
■
■
If you want to do automated mixes, this system requires a mixing console with automation, or a computer recording program
with automation.
If you mix with a hardware mixer rather than a computer, you’ll
also need some hardware effects.
Lacks the advantages of a computer recording system.
Recorder–Mixer
Pros:
■
■
■
■
■
■
■
■
Easy to use—hands on.
Simple. No need for a computer, external multitrack recorder,
or outboard effects.
Portable—good for on-location recording, but bigger than an
HD recorder.
Less likely to crash than a computer recording system.
Can export audio data to a computer for editing, or use its own
editing firmware.
All-in-one system. No external cables needed between recorder
and mixer or between mixer and effects.
Low-cost CD backup.
Automated mixing.
Cons:
■
■
■
■
Lacks the advantages of a computer recording system.
Less flexible than recording software. Not as upgradable.
However, some recorder–mixers have expansion cards that
accept downloadable plug-in effects.
May require copying data to a computer for sophisticated editing and plug-ins.
If you already have a mixer, a recorder–mixer may be more than
you need. Just get a recorder, unless you need automation.
Computer Recording System (DAW)
Pros:
■
Can be low cost if you already have a fast computer (except for
the top-level Pro Tools systems). Software effects (plug-ins) cost
much less than hardware effects.
Digital Recording CHAPTER 9
■
■
■
■
■
■
■
■
■
Flexible (due to software upgrades and downloadable plug-ins).
Everything can be “in the box.” The mixer, recorder, effects, and
CD burner are all in the computer. Fewer cables are needed.
Fast backup of files to CD or external hard drive.
Automated mixing.
When you quit a project, the computer remembers all your
mixer settings and edits. It’s easy to recall a project.
Sophisticated editing.
Can be used with your existing analog or digital mixer.
You record tracks directly to the computer. You don’t need to
copy external recordings to the computer before editing and
mastering them.
It’s easy to find parts of a song by looking at the tracks on screen.
Cons:
■
■
■
■
■
Computer may crash.
Not as “hands on” as a recorder–mixer, but a controller surface
can help with this.
Not easily portable except in laptop systems.
Requires a fast computer with a large hard drive optimized for
digital audio.
Requires a lot of computer tweaks and strong computer skills
(but this is a plus for some users).
As we’ve seen, there is a wide variety of digital recording formats.
Learn all you can about digital technology, then choose the formats
that meet your needs.
BACKUP
The hard drive on which you store your projects will eventually crash.
So it’s vital to back up your recordings’ WAV files and session files
after each session. That is, copy them to another storage medium.
Some backup media are CD-R, CD-RW, hard drive (internal or
external FireWire), DVD-R, and DVD-RW. Some engineers prefer to
archive audio programs on analog tape as well as digital media.
Clone your system drive using a program such as Casper. If your
hard drive crashes during a session, you can replace it with the clone
drive, reboot, and be back in business.
193
This page intentionally left blank
CHAPTER 10
Effects and Signal
Processors
With effects, your mix sounds more like a real “production” and less
like a bland home recording. You might simulate a concert hall with
reverb. Put a guitar in space with stereo chorus. Make a kick drum
punchy by adding compression. Used on all pop-music records,
effects can enhance plain tracks by adding spaciousness and excitement. They are essential if you want to produce a commercial sound.
But many jazz, folk, and classical groups sound fine without any
effects.
This chapter describes the most popular signal processors and effects,
and suggests how to use them.
Effects are available both as hardware and software (called plug-ins).
To add a hardware effect to a track, you feed the track’s signal from
your mixer’s aux send to an effects device, or signal processor (Figure
10.1). It modifies the signal in a controlled way. Then the modified
signal returns to your mixer, where it blends with the dry, unprocessed signal.
SOFTWARE EFFECTS (PLUG-INS)
Most recording programs include plug-ins: software effects that you
control on your computer screen. Each effect is an algorithm (small
program) that runs either in your computer’s CPU or in a DSP card.
Some plug-ins come already installed with the recording software;
others you can download or purchase on CD, then install them on
your hard drive. Each plug-in becomes part of your recording program (called the host), and can be called up from within the host.
195
196
CHAPTER 10 Effects and Signal Processors
FIGURE 10.1
Eventide H8000FW Multichannel Effects System, an example of a signal processor.
You can use plug-ins made by your recording software company or
by others. Some manufacturers make plug-in bundles, which are a
variety of effects in a single package.
Plug-ins are the usual way to create effects in a DAW. Some DAWs
let you configure your audio interface to produce an aux-send signal,
which you feed to an external hardware processor. The processed signal returns to the interface and blends with the dry signal.
All the effects described below are available as plug-ins as well as
hardware.
EQUALIZER
Recall from Chapter 2 that an equalizer (usually in the mixer) is a
sophisticated tone control, something like the bass and treble controls in a stereo system. Equalization (EQ) lets you improve on reality: add crispness to dull cymbals or add bite to a wimpy electric
guitar. EQ also can make a track sound more natural; for instance,
remove tubbiness from a close-miked vocal.
To understand how EQ works, we need to know the meaning of a
spectrum. Each instrument or voice produces a wide range of frequencies called its spectrum—the fundamentals and harmonics.
The spectrum gives each instrument its distinctive tone quality or
timbre.
If you boost or cut certain frequencies in the spectrum, you change the
tone quality of the recorded instrument. EQ adjusts the bass, treble,
and midrange of a sound by turning up or down certain frequency
ranges; that is, it alters the frequency response. For example, a boost
(a level increase) in the range centered at 10 kHz makes percussion
sound bright and crisp. A cut at the same frequency dulls the sound.
Effects and Signal Processors CHAPTER 10
FIGURE 10.2
The effect of the bass and treble control.
Types of EQ
Equalizers range from simple to complex. The most basic type is a
bass and treble control (labeled LF EQ and HF EQ). Figure 10.2
shows its effect on frequency response. Typically, this type has up to
15 dB of boost or cut at 100 Hz (for the low-frequency EQ knob) and
at 10 kHz (for the high-frequency EQ knob).
With 3-band EQ you can boost or cut the lows, mids, and highs
at fixed frequencies (Figure 10.3). Sweepable EQ is more flexible
because you can “tune in” the exact frequency range needing adjustment (Figure 10.4). If your mixer has sweepable EQ, one knob sets
the center frequency while another sets the amount of boost or cut.
Parametric EQ lets you set the frequency, amount of boost/cut, and
bandwidth—the range of frequencies affected. Figure 10.5 shows
how a parametric equalizer varies the bandwidth of the boosted part
of the spectrum. The “Q” or quality factor of an equalizer is the center frequency divided by the bandwidth. A boost or cut with a low-Q
setting (like 1.5) affects a wide range of frequencies; a high-Q setting
(like 10) makes a narrow peak or dip.
A graphic equalizer (Figure 10.6) is usually outside the mixing console. This type has a row of slide pots that work on 5 to 31 frequency
bands. When the controls are adjusted, their positions graphically
show the resulting frequency response. Usually, a graphic equalizer
is used for monitor-speaker EQ, or is patched into a channel for
sophisticated tonal tweaking.
197
198
CHAPTER 10 Effects and Signal Processors
BOOST
BOOST
dB
CUT
FREQUENCY IN Hz
FIGURE 10.3
The effect of 3-band EQ.
FIGURE 10.4
The effect of sweepable EQ.
Equalizers can also be classified by the shape of their frequency
response. Peaking EQ shapes the response like a hill or peak when
set for a boost (Figure 10.7). With shelving EQ, the shape of the
frequency response resembles a shelf (Figure 10.8). Audio clip 27 at
Effects and Signal Processors CHAPTER 10
FIGURE 10.5
Curves that illustrate varying the bandwidth of a parametric equalizer.
FIGURE 10.6
Rane ME60S, an example of a graphic equalizer.
dB
FREQUENCY IN Hz
FIGURE 10.7
Peaking EQ at 7 k Hz.
www.elsevierdirect.com/companions/9780240811444 demonstrates various
types of EQ.
A filter causes a rolloff at the frequency extremes. It sharply rejects
(attenuates) frequencies above or below a certain frequency. Figure
10.9 shows three types of filters: lowpass, highpass, and bandpass. For
example, a 10-kHz lowpass filter (high-cut filter) removes frequencies
above 10 kHz. Its response is down 3 dB at 10 kHz and more above
199
200
CHAPTER 10 Effects and Signal Processors
dB
FREQUENCY IN Hz
FIGURE 10.8
Shelving EQ at 7 kHz.
FIGURE 10.9
Lowpass, highpass, and bandpass filters.
that. This reduces hiss-type noise without affecting tone quality as
much as a gradual treble rolloff would. A 100-Hz highpass filter (lowcut filter) attenuates frequencies below 100 Hz. Its response is down
3 dB at 100 Hz and more below that. This removes low-pitched noises
such as air handler rumble or breath pops. A 1-kHz bandpass filter
cuts frequencies above and below a frequency band centered at 1 kHz.
The crossover filter in some monitor speakers consists of lowpass,
highpass, and bandpass filters. They send the lows to the woofer,
mids to the midrange, and highs to the tweeter.
A filter is named for the steepness of its rolloff: 6 dB per octave (first
order), 12 dB per octave (second order), 18 dB per octave (third
order), and so on.
How to Use EQ
If your mixer has bass and treble controls, their frequencies are preset (usually at 100 Hz and 10 kHz). Set the EQ knob at 0 to have
Effects and Signal Processors CHAPTER 10
Table 10.1
Frequency Ranges of Musical Instruments and Voices
Instrument
Fundamentals
Harmonics
Flute
261–2349 Hz
3–8 kHz
Oboe
261–1568 Hz
2–12 kHz
Clarinet
165–1568 Hz
2–10 kHz
Bassoon
62–587 Hz
1–7 kHz
Trumpet
165–988 Hz
1–7.5 kHz
French horn
87–880 Hz
1–6 kHz
Trombone
73–587 Hz
1–7.5 kHz
Tuba
49–587 Hz
1–4 kHz
100–200 Hz
1–20 kHz
Snare drum
30–147 Hz
1–6 kHz
Cymbals
Kick drum
300–587 Hz
1–15 kHz
Violin
196–3136 Hz
4–15 kHz
Viola
131–1175 Hz
2–8.5 kHz
Cello
65–698 Hz
1–6.5 kHz
Acoustic bass
41–294 Hz
700 Hz–5 kHz
Electric bass
41–294 Hz
700 Hz–7 kHz
Acoustic guitar
82–988 Hz
1500 Hz–15 kHz
Electric guitar
82–1319 Hz
1–15 kHz (direct)
Elec. guitar amp
82–1319 Hz
1–4 kHz
Piano
28–4196 Hz
5–8 kHz
Bass (voice)
87–392 Hz
1–12 kHz
Tenor (voice)
131–494 Hz
1–12 kHz
Alto (voice)
175–698 Hz
2–12 kHz
Soprano (voice)
247–1175 Hz
2–12 kHz
no effect (flat setting). Turn it clockwise for a boost; turn it counterclockwise for a cut. If your mixer has multiple-frequency EQ or
sweepable EQ, one knob sets the frequency range and another sets
the amount of boost or cut.
Table 10.1 shows the fundamentals and harmonics of musical instruments and voices. The harmonics given represent an approximate
range. Percussion, cymbals, and muted trumpet actually have some
201
202
CHAPTER 10 Effects and Signal Processors
energy up to 80 to 100 kHz. For each instrument, turn up the lower
end of the fundamentals to get warmth and fullness. Turn down the
fundamentals if the tone is too bassy or tubby. Turn up the harmonics for presence and definition; turn down the harmonics if the tone
is too harsh or sizzly.
Here are some suggested frequencies to adjust for specific instruments.
If you want the effects described below, apply boost. If you don’t,
apply cut. Try these suggestions and accept only the sounds you like:
■
■
■
■
■
■
■
■
Bass: Full and deep at 60 to 100Hz, growl at 600 Hz, presence
at 2.5 kHz, string noise at 3 kHz and up. Cut around 200 to
500 Hz for clarity.
Electric guitar: Thumpy at 60 Hz, full at 100 Hz, puffy at 500 Hz,
presence or bite at 2 to 3 kHz, sizzly and raspy above 6 kHz.
Drums: Full at 100 to 200 Hz, wooly at 250 to 800 Hz (try cutting in that range), trashy snare at 1 to 3 kHz, attack at 5 kHz.
Try cutting toms around 600 Hz to reduce boxiness. Dull cymbals sound more sizzly and crisp with a boost at 10 to 12 kHz.
Kick drum: Full and powerful below 60 Hz, papery at 300 to
800 Hz (cut at 400 to 600 Hz for better tone), click or attack at
2 to 6 kHz.
Sax: Warm at 500 Hz, harsh at 3 kHz, key noise above 10 kHz.
Acoustic guitar: Full or thumpy at 80 Hz, presence at 5 kHz, pick
noise above 10 kHz.
Acoustic guitar pickup: To make the guitar sound more “acoustic,” try a narrow cut at 1.2 to 1.5 kHz and maybe some highfrequency cut.
Voice: Full at 100 to 200 Hz (males), full at 200 to 400 Hz
(females), honky or nasal at 500 Hz to 1 kHz, presence at 5 kHz,
sibilance (“s” and “sh “sounds) around 3 to 10 k Hz.
Example: Suppose a vocal track sounds too full or bassy. Reach
for the LF EQ knob (say, 100 Hz) and turn it down until the voice
sounds natural. Or suppose a snare drum sounds dull or muffled.
Grab the mid-frequency EQ knob, set it to 5 to 10kHz, and turn it up
until the snare sounds clear and crisp.
Set EQ to the approximate frequency range you need to work on.
Then apply full boost or cut so the effect is easily audible. Finally,
fine-tune the frequency and amount of boost or cut until the tonal
balance is the way you like it.
Effects and Signal Processors CHAPTER 10
What if an instrument sounds honky, tubby, or harsh, and you don’t
know what frequency to tweak? Set a sweepable EQ for extreme
boost. Then sweep the frequencies until you find the frequency range
matching the coloration. Cut that range by the amount that sounds
right. For example, a piano miked with the lid closed might have a
tubby coloration—maybe too much output around 300 Hz. Set your
low-frequency EQ for boost, and vary the center frequency until the
tubbiness is exaggerated. Then cut at that frequency until the piano
sounds natural.
In general, avoid excessive boost because it can distort the signal. Try
cutting the lows instead of boosting the highs. To reduce muddiness
or enhance clarity, cut 1 to 2 dB around 300 Hz—either on individual instruments or on the entire mix. Don’t boost everything at the
same frequency.
When to Use EQ
Before using EQ, try to get the desired tone quality by changing the
mic or its placement. This gives a more natural effect than EQ. Many
purists shun the use of EQ, complaining of excessive phase shift or
ringing caused by the equalizer—a “strained” sound. Instead, they
use carefully placed, high-quality microphones to get a natural tonal
balance without EQ.
The usual practice is to record flat (without EQ) and then equalize
the track during mixdown.
Sometimes the instruments need a lot of EQ to sound good. If
so, you might want to record with EQ so that the playback for the
musicians will sound good. When you play the multitrack recording through your monitor mixer, the recording may not sound right
unless the tracks are already equalized. (That’s assuming the monitor
mixer in your board has no EQ.)
Uses of EQ
Here are some applications for EQ:
Improve tone quality. The main use for EQ is to make an instrument sound better tonally. For example, you might use a
high-frequency rolloff on a singer to reduce sibilance, or on a
direct-recorded electric guitar to take the “edge” off the sound.
203
204
CHAPTER 10 Effects and Signal Processors
You could boost 100 Hz on a floor tom to get a fuller sound,
or cut around 250 Hz on a bass guitar for clarity. Cut around
100 Hz to reduce bass buildup on massed harmony vocals.
The frequency response and placement of each mic affect tone
quality as well.
Although you can set the EQ for each track when it is soloed, a
better way is to set the equalizers when the entire mix is playing. That’s because one instrument can mask or hide certain
frequencies in another instrument. For example, the cymbals
might mask the “s” sounds in the vocal, making the vocal
sound dull—even though it might sound fine when soloed.
Create an effect. Extreme EQ reduces fidelity, but it also can
make interesting sound effects. Sharply rolling off the lows and
highs on a voice, for instance, gives it a “telephone” sound.
A 1-kHz bandpass filter does the same thing. To make a mono
keyboard track sound stereo, send it to two mixer channels.
Boost lows and cut highs in one channel panned left; cut lows
and boost highs in the other channel panned right.
Reduce noise and leakage. You can reduce low-frequency
noises—bass leakage, air-conditioner rumble, mic-stand
thumps—by turning down the lows below the fundamentalfrequency range of the instrument you’re recording. This information is shown in Table 10.1.
For example, a fiddle’s lowest frequency is about 200 Hz, so
you’d use a low-cut filter (highpass filter) set to 200 Hz (if possible). This low-cut filter won’t change the fiddle’s tone quality because the filtered-out frequencies are below the fiddle’s
lowest frequency. Similarly, a kick drum has little or no output
above 9 kHz, so you can filter out highs above 9 kHz on the
kick drum to reduce cymbal leakage. Filtering out frequencies
below 100 Hz on most instruments reduces air-conditioning
rumble and breath pops. Try rolling off the lows on audience mics to prevent muddy bass. To reduce hum, set a parametric EQ for a 24-dB cut, Q of 30, at these frequencies: 60,
120, and 180 Hz (in the United States) or 50, 100, and 150 Hz
(in Europe).
Compensate for the Fletcher-Munson effect. As discovered by
Fletcher and Munson, the ear is less sensitive to bass and treble
at low volumes than at high volumes. So, when you record a
very loud instrument and play it back at a lower level, it might
Effects and Signal Processors CHAPTER 10
lack bass and treble. To restore these, you may need to boost
the lows (around 100 Hz) and the highs (around 4 kHz) when
recording loud rock groups. The louder the group, the more
boost you need. It also helps to use cardioid mics with proximity effect (for bass boost) and a presence peak (for treble boost).
Make a pleasing blend. If you mix two instruments that sound
alike, such as lead guitar and rhythm guitar, they tend to mush
together—it’s hard to tell what each is playing. You can make
them more distinct by equalizing them differently. For example, make the lead guitar edgy by boosting 3 kHz, and make
the rhythm guitar mellow by cutting 3 kHz. Then you’ll hear
a more pleasing blend and a clearer mix. The same philosophy applies to bass guitar and kick drum. Because they occupy
about the same low-frequency range, they tend to mask or
cover each other. To make them distinct, either fatten the bass
and thin out the kick a little, or vice versa. The idea is to give
each instrument its own space in the frequency spectrum; for
example, the bass fills in the lows, synth chords emphasize
mid-bass, lead guitar adds edge in the upper mids, and cymbals add sparkle in the highs.
Compensate for mic placement. Sometimes you’re forced to
mike very close to reject background sounds and leakage. But a
close mic emphasizes the part of the instrument that the mic is
near. This gives a colored tone quality, but EQ can partly compensate for it. Suppose you had to record an acoustic guitar
with a mic near the sound hole. The guitar track will sound
bassy because the sound hole radiates strong low frequencies.
But you can turn down the lows on your mixer to restore a natural tonal balance.
This use of EQ can save the day by fixing poorly recorded tracks
in live concert recordings. During a concert, the stage monitors might be blaring into your recording/PA microphones, so
you’re forced to mike close to reject monitor leakage and feedback. This close placement, or the monitor leakage itself, can
give the recording an unnatural tone quality. In this case, EQ is
the only way to get usable tracks.
“Re-mix” a single track. If a track contains two different instruments, sometimes you can change the mix within that track
by using EQ. Imagine a track that has both bass and synth.
By using LF EQ, you can bring the bass up or down without
205
206
CHAPTER 10 Effects and Signal Processors
affecting the synth very much. Mixing with EQ is more effective when the two instruments are far apart in their frequency
ranges.
Improve the tonal balance of an entire mix. During mastering,
you can EQ the stereo mix of each song to make it better, to
make the songs on an album sound more similar, or to make
the album sound more like commercial albums. An effective
tool for this purpose is Harmonic Balancer (www.har-bal.com).
Whenever you record, the ideal situation is to use the right mic
in the right position, and in a good-sounding room. Then you
don’t need or want EQ. Otherwise, though, your recordings
will sound better with EQ than without it.
COMPRESSOR
A compressor acts like an automatic volume control, turning down
the volume when the signal gets too loud. Here’s why it’s necessary.
Suppose you’re recording a female vocalist. Sometimes she sings too
softly and gets buried in the mix; other times she hits loud notes and
blasts the listener. Or she may move toward and away from the mic
while singing, so that her average recording level changes.
To control this problem, you can ride gain—turn her down when
she gets too loud; turn her up when she gets too quiet. But it’s
hard to anticipate these changes. You might prefer to use a compressor, which does the same thing automatically. It reduces the
gain (amplification) when the input signal exceeds a preset level
(called the threshold). The greater the input level, the less the
gain. As a result, loud notes are made softer, so the dynamic range
is reduced (Figure 10.10). Play audio clip 28 at www.elsevierdirect
.com/companions/97802408114444.
Compression keeps the level of vocals or instruments more constant, so they are easier to hear throughout the mix, and it prevents
loud notes that might clip. Also, it can be used for special effects—to
make drums sound fatter, or to increase the sustain on a bass guitar.
In pro studios, compression is used almost always on vocals; often
on bass guitar, kick drum, and acoustic guitar; and sometimes on
other instruments.
Effects and Signal Processors CHAPTER 10
FIGURE 10.10
Compression.
Doesn’t compression rob the music of its expressive dynamics? Yes,
if overdone. But a vocal that gets too loud and soft is annoying. You
need to tame it with a compressor. Even then, you can tell when the
vocalist is singing loudly by the tone of the voice. It also helps to
compress the bass and kick drum to ensure a uniform, driving beat.
You can avoid vocal compression if the singer uses proper mic technique. He or she should back away from the mic on loud notes, and
come in close on soft notes. To tell whether you need a compressor,
listen to your finished mix. If you can understand all the words, and
no notes are too loud, omit the compressor.
Using a Compressor
Normally, you compress individual tracks or instruments, not the
entire mix. You want to compress only the stuff that needs it. To
compress a stereo mix, you need a 2-channel compressor with a stereo link, which keeps the left–right balance from changing during
compression.
Multiband compression (covered later) is usually a better choice for
compressing the stereo mix.
Let’s describe the controls on the compressor. Some compressors
have few controls; most of their settings are preset at the factory.
Figure 10.11 shows a compressor plug-in.
COMPRESSION RATIO OR SLOPE
This is the ratio of the change in input level to the change in output
level. For example, a 2:1 ratio means that for every 2 dB change in input
level, the output of the compressor changes 1 dB. A 20-dB change in
input level results in a 10-dB change in the output, and so on.
207
CHAPTER 10 Effects and Signal Processors
Output level
Knee
Threshold
208
Input level
FIGURE 10.11
Sonitus Compressor, an example of a compressor plug-in. (The graph labels are by the author.)
Typical ratio settings are 2:1 to 4:1. A “soft knee” or “over easy” characteristic is a low compression ratio for low-level signals and a high
ratio for high-level signals. Some manufacturers say that this characteristic sounds more natural than a fixed compression ratio.
THRESHOLD
This is the input level above which compression occurs. Set the
threshold high (about –5 dB) to compress only the loudest notes; set
it low (⫺10 or ⫺20 dB) to compress a broader range of notes. A setting of ⫺10 is typical. If the compressor has a fixed threshold, adjust
the amount of compression with the input level control.
GAIN REDUCTION
This is the number of dB that the gain is reduced by the compressor.
It varies with the input level. You set the ratio and threshold controls
Effects and Signal Processors CHAPTER 10
so that the gain is reduced on loud notes by an amount that sounds
right. The amount of gain reduction shows up on a meter—3 to
10 dB is typical.
COMPRESSION GRAPH
Most compressor plug-ins display a compression graph of output level versus input level in dB (Figure 10.11). As the input level
increases from zero up to the threshold level, the graph is an upwardsloping straight line in which the change in output level matches the
change in input level. Where the input level is above the threshold,
the graph transitions to a flatter slope where compression occurs:
the change in output level is less than the change in input level. The
transition point is called the knee of the compression curve. The
transition from no compression to compression can be abrupt (hard
knee) or gradual (soft knee).
The slope of the compressed part of the curve (the part on the right)
indicates the compression ratio. The more horizontal this part is, the
greater the ratio. A horizontal line indicates limiting (explained later
in this chapter under the heading Limiter). As you vary the compression ratio and threshold, the compression curve changes accordingly.
ATTACK TIME
This is how fast the compressor reduces the gain when it’s hit by a
musical attack. Typical attack times range from 0.25 to 10 msec. Some
compressors adjust the attack time automatically to suit the music;
others have a factory-set attack time. The longer the attack time, the
larger the peaks that are passed before gain reduction occurs. So, a
long attack time sounds punchy; a short attack time reduces punch
by softening the attack.
RELEASE TIME
This is how fast the gain returns to normal after a loud passage ends.
It’s the time the compressor takes to reach 63% of its normal gain.
You can set the release time from about 50 msec to several seconds.
One-half second to 0.2 second is typical. For bass instruments, the
release time must be longer than about 0.4 second to prevent harmonic distortion.
209
CHAPTER 10 Effects and Signal Processors
Short release times make the compressor follow rapid volume
changes in the music, and keep the average level higher. But because
the noise rises along with the gain, short release times can give a
pumping or breathing sound. Long release times sound more natural. If the release time is too long, though, a loud passage will reduce
the gain during a subsequent quiet passage. In some units, the
release time varies automatically, or is factory-set to a useful value.
Some compressors disable the attack and release settings when
the compressor is set to RMS or average mode. Those settings are
adjusted automatically.
Figure 10.12 shows the effects of compressor attack time and release
time on the envelope of a waveform. The gray block represents a
musical note fed into the compressor. As Figure 10.12 shows, a long
OUT
IN
Fast attack time
Level
Note’s envelope
Level
210
Onset of compression
Time
Time
Slow attack time
Onset of compression
Fast release time
Loud note
Quiet note
Slow release time
Loud note
Quiet note
End of release time
Release time matches
the note’s decay time
FIGURE 10.12
Effects of attack time and release time on the envelope of a note.
Effects and Signal Processors CHAPTER 10
attack time tends to increase the attack portion of the envelope, giving a sound with a sharper attack or “edge.” A short release time
tends to bring up the level in between notes, giving a louder but
more fatiguing sound. A release time similar to the note’s decay time
increases sustain.
OUTPUT-LEVEL CONTROL
Also called make-up gain, this control is used to increase the output
level of the compressor by the amount of gain reduction. For example, if a compressor is causing 6 dB of level reduction, set the makeup gain to 6 dB to achieve unity gain. This also brings up the quiet
parts of the track by 6 dB. Some compressors keep the output level
constant when other controls are varied.
Spend some free time playing with all the settings so you learn how
they affect the sound. Play various instruments and vocals through a
compressor, vary the settings, and take notes on what you hear.
Some compressors have a side chain. This is a pair of in/out jacks for
connecting an equalizer. To compress only the sibilant sounds on a
vocal track, boost the side-chain EQ around 3 to 10 kHz. To compress only the breath pops on a vocal track, boost the side-chain EQ
around 20 Hz.
A multiband or split-frequency compressor divides the audio band
into three to five bands (bass, mids, treble) and compresses each
band separately. That way, the compressor can squash a loud bass
note, or soften “s” sounds, without bringing down the overall level.
Multiband compression is sometimes applied to the final mix of
each song during mixdown or mastering.
Connecting a Compressor
To compress one track in DAW, select a compressor plug-in for that
track. Don’t use an aux send for compression. Connect a hardware
compressor in line with the signal you want to compress in one of
the following ways:
■
To compress one instrument or voice while recording: Locate
the input module of the instrument you want to compress. On
the back of that module, connect the insert send jack to compressor in; connect compressor out to the insert return jack.
211
212
CHAPTER 10 Effects and Signal Processors
■
■
(Chapter 11 explains these terms.) Or, take a signal from the
input module’s direct out. Feed that into the compressor, and
feed the compressor output to the recorder track input.
To compress a group of instruments while recording: Locate
the bus output of the instruments you want to compress. Go
from bus out to compressor in, and go from compressor out to
recorder-track in. If the bus has insert jacks, you could connect
to them instead.
To compress one track during mixdown: Go from track out to
compressor in, and go from compressor out to mixer channel
in. Or locate the mixer input module for that track, then find
the insert send and return jacks in that module. (There might
be a single insert jack with send and return terminals.) Connect
the insert send to compressor in; connect compressor out to
insert return.
Suggested “Ballpark” Compressor Settings
■
■
■
■
■
Vocals: Ratio 2:1 to 3:1, fast attack, 1/2 second release, set
threshold for 3 to 6 dB of gain reduction. Singers with extreme
dynamic range might need 12 dB of gain reduction and a ratio
of 4:1.
Bass and drums: Ratio 4:1, slow release, set threshold for 3 to
6 dB of gain reduction on loud bass “pops.” Adjust attack time
depending on how much you want to soften the attack. Short
attack time ⫽ soft attack, long attack time ⫽ loud attack.
Electric guitar: 4:1 to 8:1 ratio, 10-dB gain reduction, 400 msec
release.
To reduce breath pops: Use a multiband compressor. Enable
only the lowest frequency band. Try these settings: ratio 30:1,
upper frequency 100 Hz, make-up gain 0 dB, attack 1 msec,
release 100 msec, threshold ⫺18 dB. Experiment with the threshold setting.
To reduce sibilance (de-ess mode): First, set an equalizer for a
narrow boost around 6 kHz. While playing the vocal track, sweep
the frequency up and down until you find the frequency area
where sibilance (“s” and “sh” sounds) is exaggerated the most.
Note that frequency. Remove the equalizer and insert a multiband compressor. Disable all bands except a band from about
Effects and Signal Processors CHAPTER 10
■
3 to 10kHz. Try to set up a narrower band around the frequency
you noted earlier. Suggested settings: ratio 10:1, attack 1 msec,
release 100 msec, threshold ⫺32 dB. Experiment with the threshold setting to get the desired amount of de-essing.
To compress the stereo mix: Try these settings: 2:1, soft knee,
attack 20 msec, release 200 msec. Set the threshold to get 5 to
10 dB of gain reduction. If you can hear the compressor squashing the sound, back off the gain reduction unless that’s the
sound you want.
Should you compress while tracking or mixing? If you compress
while tracking, it will be difficult or impossible to change the amount
of compression during mixdown. If you compress tracks during mixdown, you can change the settings at will.
When applying low-cut EQ to a track, insert the EQ before the compressor. Often there is too much bass on a track, and that extra bass
will trigger the compressor unless you EQ it out first. When you
apply a boost, put the EQ after the compressor so that the boost isn’t
compressed.
LIMITER
A limiter keeps signal peaks from exceeding a preset level. While a compressor reduces the overall dynamic range of the music, a limiter affects
only the highest peaks (Figure 10.13). To act on these rapid peaks, limiters have a very fast attack time—1 microsecond to 1 millisecond. The
compression ratio in a limiter is very high—10:1 or greater—and the
threshold is set high, say at 0 dB. For input levels up to 0 dB, the output
level matches the input. For input levels above 0 dB, the output level
stays at 0 dB. This prevents overload in the device following the limiter.
FIGURE 10.13
Limiting.
213
214
CHAPTER 10 Effects and Signal Processors
A compressor/limiter carries out both of the functions in its name.
It compresses the average signal levels over a wide range, and limits peaks to prevent overload. It has two thresholds: one low for the
compressor and one high for the limiter.
Limiters can be used to prevent recorder overload during field recording, or to prevent PA power amps from clipping. When you master a
program of several mixed songs in a DAW, you might use limiting
to reduce the level of signal peaks in the program. Set the threshold
about 6 dB below the highest peak level. Then apply normalization,
which raises the level of the entire program until the highest peak
in the program reaches maximum level. Limiting and normalization
create a louder program on your finished CD without compressing
the music’s dynamics.
A limiter plug-in with the “look-ahead” feature checks the upcoming
audio for peaks. Audio enters a look-ahead buffer, and the limiter
measures that audio signal and reacts quickly to reduce the incoming peaks.
NOISE GATE
A noise gate (expander) acts like an on-off switch that removes noises
during pauses in an audio signal. It reduces the gain when the input
level falls below a preset threshold. That is, when an instrument
stops playing for a moment, the noise gate drops the volume, which
removes any noise and leakage during the pause (Figure 10.14).
Note: The gate doesn’t remove noise while the instrument is playing.
Where is it used? The noise gate helps to clean up drum tracks by
removing leakage between beats. It can shorten the decay time of the
FIGURE 10.14
Gating.
Effects and Signal Processors CHAPTER 10
drums, giving a very tight sound. If you’re recording a noisy guitar
amp, try a gate to cut out the buzz and hiss between phrases.
How do you use a noise gate? Patch it between a recorder-track output and a mixer line input, or use a gate plug-in in a DAW. Solo the
track that you want to gate. Set the gate’s threshold so that noise and
leakage go away during pauses. If the gate chops off each note, the
threshold is set too high—turn it down. Set the release time as short as
possible, but long enough so that you hear the entire note (or tom hit)
before the gate cuts off the sound. To fix a boomy kick drum, adjust
the threshold until the kick sounds as “tight” as you want. That is, use
the gate to shorten the decay portion of the kick-drum’s envelope.
Excellent recordings can be made without gating. But if you want
a tighter sound, gates come in handy. Some signal processors have
compression, limiting, and noise gating in a single package.
Some gates have a side-chain input or key input. It’s an input for
an external signal that controls the gating action. The control signal
triggers the output of the gate’s main audio path. For example, you
could feed a bass guitar through the noise gate, and gate the bass
with a kick-drum signal fed into the side chain. Then the bass will
follow the kick drum’s envelope.
DELAY: ECHO, DOUBLING, CHORUS, AND FLANGING
A digital delay (or a delay plug-in) takes an input signal, holds it
in memory, then plays it back after a short delay—about 1 msec to
1 second (Figure 10.15). Delay is the time interval between the input
signal and its repetition at the output of the delay device.
If you listen to the delayed signal by itself, it sounds the same as the
undelayed (dry) signal. But if you combine the delayed and dry signals, you may hear two distinct sounds: the signal and its repetition.
FIGURE 10.15
Delaying the signal.
215
216
CHAPTER 10 Effects and Signal Processors
By delaying a signal, a processor can create several effects such as
echo, repeating echo, doubling, chorus, and flanging.
Echo
If the delay is about 50 msec to 1 second, the delayed repetition of
a sound is called an echo. This is shown in Figure 10.16 by the two
pulses. Echoes occur naturally when sound waves travel to a distant
room surface, bounce off, and return later to the listener—repeating
the original sound. A delay unit can mimic this effect. Many people
use the term delay to mean echo.
In setting up a mix with echo, you want to hear both the dry sound
and its echo. You do this by creating an effects loop: from the mixer,
to the effects box, back to the mixer. Here’s how to set up echo with
a hardware mixer and effects unit:
1. On the delay unit, set the dry/wet mix control all the way to
wet or 100% mix. Then the output of the delay unit will be
only the delayed signal.
2. Suppose you want to use aux1 as the echo control. Connect
aux1 send to delay unit IN. Connect delay unit OUT to Bus 1
and 2 IN (or to the effects-return jacks).
3. Find the mixer module for the instrument you want to add
echo to.
4. Assign the instrument to busses 1 and 2. Monitor busses 1 and 2.
5. Find the knobs labeled Bus 1 IN and Bus 2 IN. They might be
called Aux Return or Effects Return. Turn them up to 0, about
three-fourths of the way up.
6. Turn up the aux1 send knob, and there’s your echo.
FIGURE 10.16
Echo.
Effects and Signal Processors CHAPTER 10
The delayed sound mixes with the dry sound in busses 1 and 2. You
hear both sounds, which together make an echo. Each aux knob controls the amount of echo on each track, while the effect-return knobs
control the overall amount of echo on all tracks that are feeding the
echo unit.
To set up echo in a DAW, follow this procedure:
1. Create or use a stereo aux bus that has an Echo or Delay plugin enabled.
2. In a track that you want to have echo, enable an aux send to
that bus.
3. Open the Echo plug-in. Set its dry/wet mix control all the way
to wet or 100% mix.
4. In the track that you want to have echo, gradually turn up the
aux-send until you hear the desired amount of echo. In the
Echo plug-in, adjust the delay parameters for the desired effect.
SLAP ECHO
A delay from about 100 to 130 msec is called a slap echo or slapback
echo. It was often used in 1950s rock ‘n’ roll and rockabilly tunes,
and still is used today.
REPEATING ECHO
Most delay units can be made to feed the output signal back into the
input, internally. Then the signal is re-delayed many times. This creates a repeating echo—several echoes that are evenly spaced in time
(Figure 10.17, and audio clips 29 and 30 at www.elsevierdirect.com/
companions/9780240811444). The regeneration (feedback) control
sets the number of repeats.
FIGURE 10.17
Repeating echo.
217
218
CHAPTER 10 Effects and Signal Processors
Repeating echo is most musical if you set the delay time to create an
echo rhythm that fits the tempo of the song. The formula is
Delay in seconds ⫽ 60/tempo
So if the tempo is 120 bpm, the delay is 0.5 sec (500 msec). That’s
one echo per quarter note. Use half that delay to get one echo per
eighth note. Use one-third that delay for triplets. A slow repeating
echo—0.5 second between repeats; for example, gives an outer-space
or haunted-house effect. It often sounds good on lead vocal in a
ballad.
Doubling
If you set the delay around 23 to 30 msec, the effect is called doubling or automatic double tracking (ADT). It gives an instrument
or voice a fuller sound, especially if the dry and delayed signals are
panned to opposite sides. The short delays used in doubling sound
like early sound reflections in a studio, so they add some “air” or
ambience.
Doubling a vocal can be done without a delay unit. Record a vocal
part, then overdub another performance of the same vocal part. Mix
the parts, pan them both to center, or pan them left and right.
Chorus
This is a wavy or shimmering effect. The delay is 15 to 35 msec, and
the delay varies at a slow rate. Sweeping the delay time causes the
delayed signal to bend up and down in pitch, or to detune. When
you combine the detuned signal with the original signal, you get
chorusing.
STEREO CHORUS
This is a beautiful effect. In one channel, the delayed signal is combined with the dry signal in the same polarity. In the other channel,
the delayed signal is inverted in polarity, then combined with the dry
signal. Thus, the right channel has a series of peaks in the frequency
response where the left channel has dips, and vice versa. The delay is
slowly varied or modulated. Hear a demonstration with audio clip 33 at
www.elsevierdirect.com/companions/9780240811444.
Effects and Signal Processors CHAPTER 10
BASS CHORUS
This is chorus with a highpass filter so that low frequencies aren’t
chorused, but higher harmonics are. It gives an ethereal quality to
the bass guitar.
Flanging
If you set the delay around 0 to 20 msec, you usually can’t resolve
the direct and delayed signals into two separate sounds. Instead, you
hear a single sound with a strange frequency response. The direct and
delayed signals combine and have phase interference, which puts a
series of peaks and dips in the frequency response. This is called a
comb-filter effect (Figure 10.18). It gives a very colored, filtered tone
quality. The shorter the delay, the farther apart the peaks and dips
are spaced in frequency.
The flanging effect varies or sweeps the delay between about 0 and
20 msec. This makes the comb-filter nulls sweep up and down the
spectrum. As a result, the sound is hollow, swishing, and ethereal,
as if the music were playing through a pipe. Flanging is easiest to
hear with broadband signals such as cymbals but can be used on any
instrument, even voices. Hear a demonstration with audio clip 34 at
www.elsevierdirect.com/companions/9780240811444.
Some examples of flanging are on many Jimi Hendrix records, and
on the oldies “Itchycoo Park” by the Small Faces and “Listen to the
Music” by the Doobie Brothers. The first use of flanging was on “The
Big Hurt” sung by Toni Fisher.
Positive flanging refers to flanging in which the delayed signal is
the same polarity as the direct signal (Figure 10.18). With negative flanging, the delayed signal is opposite in polarity to the direct
FIGURE 10.18
Flanging (or positive flanging).
219
220
CHAPTER 10 Effects and Signal Processors
FIGURE 10.19
Negative flanging.
signal, which makes a stronger effect. The low frequencies are canceled (the bass rolls off), and the “knee” of the bass rolloff moves up
and down the spectrum as the delay is varied. The high frequencies
are still comb-filtered (Figure 10.19). Negative flanging makes the
music sound like it’s turning inside out.
When the flanger feeds some of the output signal back into the
input, the peaks and dips get bigger. It’s a powerful “science fiction”
effect called resonant flanging.
REVERBERATION
This effect adds a sense of room acoustics, ambience, or space to
instruments and voices. To know how it works, we need to understand how reverb happens in a real room. Natural reverberation in
a room is a series of multiple sound reflections that make the original sound persist and gradually die away or decay. These reflections
tell the ear that you’re listening in a large or hard-surfaced room. For
example, reverberation is the sound you hear just after you shout in
an empty gymnasium.
A reverb effect simulates the sound of a room—a club, auditorium,
or concert hall—by generating random multiple echoes that are too
numerous and rapid for the ear to resolve (Figure 10.20). Digital
reverb is available either in a dedicated reverb unit, as part of a
multi-effects processor, or as a plug-in.
The most natural sounding digital reverb is a sampling reverb or convolution reverb, which creates the reverb from impulse-response samples (WAVE files) of real acoustic spaces, rather than from algorithms.
One convolution reverb plug-in is SIR at www.knufinke.de/sir/. Free
impulse-response samples are at www.noisevault.com and www.cksde
.com/p_6_250.htm.
Effects and Signal Processors CHAPTER 10
FIGURE 10.20
Reverberation.
Reverb and echo aren’t the same thing. Echo is a repetition of a
sound (HELLO hello hello); reverb is a smooth decay of sound
(HELLO-OO-oo-oo).
Multichannel digital reverbs are available for surround sound, both
as hardware and software. Examples include: Eventide’s Orville,
Sony’s DRE-S777, TC Electronics’ System 6000, Lexicon’s 960L, and
Kind of Loud’s RealVerb 5.1 Pro Tools plug-in. Surround reverb plugins for Steinberg’s Nuendo platform include Steinberg’s Surround
Edition plug-in bundle and TC Works SurroundVerb plug-in.
Reverb Parameters
Here are some controls in a reverb unit or plug-in (Figure 10.21):
■
■
■
Reverb Time (RT60): The time it takes for reverberation to
decay 60 dB below its original level. Set it long (1–1/2 to 2 seconds) to simulate a large room; set it short (under 1 second) to
simulate a small room. Generally you use short reverbs (or no
reverb) for fast songs, and long reverbs for slow songs.
Pre-delay (pre-reverb delay): A short delay (10 to 100 msec)
before the onset of reverb to simulate the delay that happens
in real rooms before reverb starts. The longer the pre-delay, the
bigger the room sounds. Using pre-delay on an instrument’s
reverb often helps to clarify the sound by removing the onset of
reverb from the direct sound of the instrument. If your reverb
unit doesn’t have pre-delay, you can create it by connecting a
delay unit between your mixer’s aux-send and the reverb input.
If your reverb plug-in lacks pre-delay, insert a delay plug-in
before the reverb plug-in. Audio clip 31 at www.elsevierdirect.com/
companions/9780240811444 demonstrates reverb: short reverb time,
long reverb time, and pre-delay.
Density or diffusion: A high density setting produces many
echoes spaced close together. It gives a smooth decay but
221
222
CHAPTER 10 Effects and Signal Processors
FIGURE 10.21
Lexicon Pantheon Reverb plug-in.
■
■
increases the load on the CPU. A low-density setting produces
fewer echoes spaced farther apart, and may be adequate for
vocals, synth pads, and organ. Use high density for percussive
sounds to prevent grainy-sounding reverb.
Damping: Adjusts the reverb time or decay at high frequencies.
Set the damping frequency high (say, 7 kHz) to simulate a hardsurfaced room; set it low (say, 2 kHz) to simulate a soft-surfaced
room. The latter is also called a “warm room” reverb.
Presets: Factory-supplied reverb settings of small rooms, auditoriums, halls, and so on. A plate reverb setting duplicates the
bright sound of a metal-foil plate, which used to be the most
popular type of reverb in pro studios. Unnatural effects are
available, such as nonlinear decay, reverse reverb that builds up
before decaying, or gated reverb. With gated reverb, the reverb
cuts off suddenly shortly after a note is hit. In the 1980s it was
often used on a snare drum. A good example is the oldie “You
Can Call Me Al” on Paul Simon’s album Graceland.
Effects and Signal Processors CHAPTER 10
Reverb Connections
To connect a reverb unit to a hardware mixer, connect a cable from
the mixer aux-send to the reverb input. Connect a cable (two for stereo) from the reverb outputs to the mixer aux returns (effects returns
or bus inputs). Set the mix control on the reverb unit all the way to
wet or reverb. Turn the mixer’s aux-return or bus-in knobs (if any)
about two-thirds of the way up, and adjust the amount of reverb on
each track with the aux-send knobs. Try to get an overall reverb-send
level near 0 on the meter; then fine-tune the aux return level for the
desired amount of reverb.
To enable reverb in a DAW, use the same procedure as for setting up
echo, but choose a reverb plug-in.
PREVERB
Preverb is reverb that precedes a note rather than follows it. The
reverb starts from silence and builds up to a note’s attack (audio clip
32 at www.elsevierdirect.com/companions/9780240811444). When used
on a snare drum, preverb gives a whip-cracking kind of sound, like
“shSHK!”
Here’s how to add preverb to a snare drum track in a DAW:
1. Set up an aux1 bus with reverb. Set the reverb all the way to
wet or 100% mix.
2. Solo the drum track.
3. On the drum track, turn up the aux1 send and set it to prefader. Find a good snare hit and play it.
4. Turn down the drum-track fader and play the snare hit. You
should hear only the reverb from the aux1 bus.
5. Highlight the snare hit up until the next hit. Export or save
the mix (the snare hit) as Drum reverb.wav.
6. Select a blank track and call it Drum Reverb. Import Drum
reverb.wav into that track.
7. Select the drum-reverb clip, then select the Reverse processing
in your DAW. This reverses the drum-reverb clip.
8. On the original snare-drum track, reset the aux1 send to postfader and turn it back down. Turn up the track fader where it
used to be.
223
224
CHAPTER 10 Effects and Signal Processors
9. Slide in time the reversed drum-reverb clip so that it ends just
as a snare-drum hit starts (check the waveforms).
10. Play the reversed drum-reverb track along with the snaredrum track. You should hear preverb.
Some signal processors have a reverse reverb effect in which the
reverb comes after the note that produced it, but builds up before
it fades out. This isn’t quite the same as preverb. Reverse reverb can
upset the musical timing; preverb doesn’t.
ENHANCER
If a track or a mix sounds dull and muffled, you can run it through
an enhancer to add brilliance and clarity. An enhancer works by adding slight distortion (as in the Aphex Aural Exciter).
OCTAVE DIVIDER
This unit takes a signal from a bass guitar and provides deep, growling bass notes one or two octaves below the pitch of the bass guitar. It does this by dividing the incoming frequency by 2 or 4: If you
put 82 Hz in, you get 41 Hz out. Some MIDI sound modules have
bass patches with extra-deep sound, and 5-string bass guitars have an
extra string tuned especially low.
HARMONIZER
Basically a delay unit with delay modulation, a harmonizer makes a
variety of pitch-shifting effects. It can create harmonies, change pitch
without changing the duration of the program, change duration
without changing pitch, and many other oddities. You’ve heard harmonizers on radio-station spots when the announcer’s voice sounds
like a Munchkin or Darth Vader. Play audio clip 35 at www.elsevierdirect
.com/companions/9780240811444.
VOCAL PROCESSOR
This device or plug-in can affect the vocal’s inflection, add growls or
whispers, correct the pitch, add vibrato, make the voice more-or-less
nasal or chesty, and so on. The latest vocal formant-corrected pitchshifters maintain the voice formant structure when they shift pitch;
Effects and Signal Processors CHAPTER 10
this prevents the “chipmunk” effect. An example includes TC-Helicon
VoiceModeler plug-in.
Another type of vocal processor is called a channel strip. It includes
a high-quality mic preamp or two, plus EQ, compression, gating,
de-esser, and perhaps some tube saturation distortion. An example is
Focusrite VoiceMaster Pro.
PITCH CORRECTION
This plug-in provides automatic or manual pitch correction of a single track (but not of chords). In automatic mode, it corrects flat or
sharp notes by changing their pitch to match a musical scale of your
choice. In manual mode, you see a graph of the notes’ pitches on
your monitor screen, and slide certain notes up or down to correct
their pitch. Manual mode is less obvious than automatic. You also
can use this effect as a “robotic” effect where the sung notes change
pitch in a step-wise, jerky way rather than smoothly. Some pitchcorrection plug-ins are Antares Auto-Tune, Celemony Melodyne, TC
Helicon Intonator, and Roland’s V-Vocal plug-in for Cakewalk Sonar
Producer.
Melodyne Direct Note Access lets you edit individual notes within
chords. See www.celemony.com.
TUBE PROCESSOR
This device or plug-in uses a vacuum tube or a simulation of one.
Tubes have euphonic even-order harmonic distortion, which is
claimed to add “richness” or “warmth” when the tube distorts (audio
clip 36 at www.elsevierdirect.com/companions/9780240811444). There
are tube mics, tube mic preamps, tube compressors, and stand-alone
tube processors.
ROTARY SPEAKER SIMULATOR
This effect simulates the sound of a Leslie organ speaker, which plays
music through rotating horns. It’s a complex sound effect of pitch
shifting, tremolo, and phase shifting. The speed and depth of the
effect are adjustable.
225
226
CHAPTER 10 Effects and Signal Processors
ANALOG TAPE SIMULATOR
Analog tape saturation is mainly third-harmonic distortion and
compression. An analog tape simulator adds this distortion to
digital recordings in an attempt to smear or warm up the sound
in a pleasant way (Audio clip 37 at www.elsevierdirect.com/companions/
9780240811444).
SPATIAL PROCESSOR
Spatial processors enhance the stereo imaging or spatial aspects of
a mix heard over two speakers. Some units have joystick-type pan
pots, which move the image of each track anywhere around the listener. Other units make the stereo stage wider, so that images can
be placed to the left of the left speaker, and to the right of the right
speaker. The listener might hear images toward the sides of the listening room. In 5.1 surround systems, this spatial processing is done
by surround panning and surround reverbs.
MICROPHONE MODELER
Antares and Roland offer a microphone modeler or simulator. You
tell it which mic you’re using and which mic you want it to sound
like. A wide variety of vintage and current mic simulations are available. Mic modeling comes in three forms: hardware device, plug-in,
and firmware (programmed into a chip) in a recorder–mixer.
GUITAR AMPLIFIER MODELER
Another simulator takes the sound of a direct-recorded guitar, and
makes the guitar sound like it is played through a guitar amp (audio
clip 38 at www.elsevierdirect.com/companions/9780240811444). Several
amp models can be simulated, as well as effects, tone, drive, the mic
used to pick up the amp, and the mic’s position.
Two hardware examples of amp modelers are the Line 6 Pod and the
Johnson J Station. Amp Farm is a guitar modeling plug-in for Pro
Tools. Roland’s DAWs offer COSM mic modeling and guitar-amp
modeling.
Guitar processors or guitar stomp boxes can be used on any instrument or vocal to add distortion.
Effects and Signal Processors CHAPTER 10
FIGURE 10.22
CamelCrusher distortion plug-in.
DISTORTION
Some plug-ins add intentional distortion of various types to generally “shred” the sound for a low-fi effect. Two examples are Izotope
Trash and Camel Audio CamelCrusher (Figure 10.22).
DE-CLICK AND DE-NOISE
Also called “Audio restoration programs,” these are plug-ins—or
stand-alone programs—that can remove the clicks and pops from LP
records, or remove hiss and hum from noisy recordings. Examples
include: Dart Pro 24, iZotope RX, Bias SoundSoap, and Waves Native
Restoration Bundle.
SURROUND SOUND
Recent plug-ins for surround sound are surround panning, surround
reverb, and surround encoding/decoding.
227
228
CHAPTER 10 Effects and Signal Processors
MULTI-EFFECTS PROCESSOR
This provides several effects in a single device or plug-in. Some units
let you combine up to four effects in any order. Others have several channels, so you can put a different effect on different instruments. With most processors, you can edit the sounds and save them
in memory as new programs. At www.elsevierdirect.com/companions/
9780240811444 audio clips 29–38 demonstrate various effects on voice.
Some processors offer 100 or more programmable presets with MIDI
control over any parameter. For example, with some units you can
place an instrument in a simulated room, and use a MIDI controller
to continuously change the size of that room. Many signal processors can be controlled by MIDI program-change commands. You can
quickly change the type of effect, or effect parameters, by entering
certain program changes into a sequencer.
Suppose you want each tom-tom hit in a drum fill to have a different size room added to it. For example, put the high-rack tom in a
small room; put the low-rack tom in a concert hall; and put the floor
tom in a cave. To do this, first assign a different program number
(patch or preset number) to each effects parameter. You do this with
the effects device. Then, using the sequencer, punch in the appropriate program number for each note.
A MIDI program-change footswitch lets guitarists call up different
effects on MIDI signal processors. By tapping a footswitch, they can
get fuzz, flanging, wah-wah, spring reverb, and so on.
A MIDI mapper lets you control some effects parameters with any
controller. For example, vary reverb decay time with a pitch wheel, or
vary a filter with key velocity.
LOOKING BACK
We’ve come a long way with effects. Looking back over the past few
decades, each era had its own “sound” related to the effects used at
the time. The 1950s had tube distortion and slap echo; the 1960s
used fuzz, wah-wah, and flanging. Much of the early 1970s sounded
dry, and the early 1990s emphasized synthesizers, drum machines,
and gated reverb. Now vacuum tubes and acoustic instruments are
back, along with occasional low-fi (tinny, distorted, or noisy) sounds
Effects and Signal Processors CHAPTER 10
and dry vocals. Whatever effects you choose, they can enhance your
music if used with taste.
SOUND-QUALITY GLOSSARY
The sound of mic techniques, effects, and EQ in a recording can
be hard to translate into engineering terms. For example, what EQ
should you use to get a “fat” sound or a “thin” sound? The glossary
below may help. It’s based on conversations with producers, musicians, and reviewers over 30 years. Not everyone agrees on these
definitions, but they are common. This glossary doesn’t suggest the
cause of the sound quality or how to change it; that’s up to you to
determine.
AIRY—Spacious. The instruments sound like they are surrounded
by a large reflective space full of air. A pleasant amount of reverb
or early reflections. High-frequency response that extends to 15 or
20 kHz.
BALLSY or BASSY—Emphasized low frequencies below about 200 Hz.
BLOATED—Excessive mid-bass around 250 Hz. Poorly damped low
frequencies, low-frequency resonances.
BLOOM—Adequate low frequencies. Spacious. Good reproduction
of dynamics and reverberation. Early reflections or a sense of “air”
around each instrument in an orchestra.
BOOMY—Excessive bass around 125 Hz. Poorly damped low frequencies or low-frequency resonances.
BOXY—Having resonances as if the music were enclosed in a box.
Speaker cabinet diffraction or vibration. An emphasis around 250 to
600 Hz.
BREATHY—Audible breath sounds in vocals, flute, or sax. Good
high-frequency response.
BRIGHT—High-frequency emphasis. Harmonics are strong relative
to fundamentals.
BRITTLE—High-frequency peaks, or weak fundamentals. Slightly
distorted or harsh highs. Opposite of round or mellow. See Thin.
229
230
CHAPTER 10 Effects and Signal Processors
Objects that are physically thin and brittle emphasize highs over
lows when you crack them.
CHESTY—A vocal signal with a bump in the low-frequency response
around 125 to 250 Hz.
CLEAN—Free of noise, distortion, and leakage.
CLEAR—See Transparent.
CLINICAL—Too clean or analytical. Emphasized high-frequency
response, sharp transient response. Not warm.
COLORED—Having timbres that aren’t true to life. Non-flat
response, peaks, or dips.
CLOUDY—See Wooly.
CONSTRICTED—Poor reproduction of dynamics. Dynamic compression. Distortion at high levels. Also see Pinched.
CRISP—Extended high-frequency response. Like a crispy potato chip
or crisp bacon frying. Often referring to cymbals.
CRUNCH—Pleasant guitar-amp distortion.
DARK—Opposite of bright. Weak high frequencies.
DELICATE—High frequencies extending to 15 or 20 kHz without
peaks. A sweet, airy, open sound with strings or acoustic guitar.
DEPTH—A sense of closeness and farness of instruments, caused
by miking them at different distances. Good transient response that
reveals the direct/reflected sound ratio in the recording.
DETAILED—Easy to hear tiny details in the music; articulate.
Adequate high-frequency response, sharp transient response.
DISTANT—Too much leakage. Low direct-to-reverb ratio.
DRY—Without effects. Not spacious. Reverb tends toward mono
instead of spreading out. Overdamped transient response.
DULL—See Dark.
ECHOEY—Having audible echoes or reverberation.
Effects and Signal Processors CHAPTER 10
EDGY—Too much high frequencies. Trebley. Harmonics are too
strong relative to the fundamentals. When you view the waveform
on an oscilloscope, it even looks edgy or jagged, because of excessive
high frequencies. Distorted, having unwanted harmonics that add an
edge or raspiness to the sound.
EFFORTLESS—Low distortion, usually coupled with flat response.
ETCHED—Clear but verging on edgy. Emphasis around 10 kHz or
higher.
FAT—See Full and Warm. Also, a diffuse spatial effect. Also, smeared
out in time, with some reverberant decay. Also, the sound of a snare
drum tuned low.
FOCUSED—Referring to the image of a musical instrument which is
easy to localize, pinpointed, having a small spatial spread.
FORWARD—Sounding close to the listener.. Emphasis around
2 to 5 kHz.
FULL—Opposite of thin. Strong fundamentals relative to harmonics. Good low-frequency response, not necessarily extended, but with
adequate level around 100 to 300 Hz.
GENTLE—Opposite of edgy. The harmonics—highs and upper
mids—aren’t exaggerated, or may even be weak.
GLARE, GLASSY—A little less extreme than edgy. A little too bright
or trebley.
GRAINY—The music sounds like it’s segmented into little grains,
rather than flowing in one continuous piece. Not liquid or fluid. Suffering from distortion. Some early A/D converters sounded grainy, as do
current ones of inferior design. “Powdery” is finer than “grainy”!
GRUNGY—Lots of distortion.
HARD—Too much upper midrange, usually around 3 kHz. Or, good
transient response, as if the sound is hitting you hard.
HARSH—Too much upper midrange. Peaks in the frequency
response from 2 to 6 kHz. Or, excessive phase shift.
231
232
CHAPTER 10 Effects and Signal Processors
HEAVY—Good low-frequency response below about 50 Hz.
Suggesting great weight or power, like a diesel locomotive or thunder.
HOLLOW—See Honky. Or, too much reverberation, a mid-frequency
dip, or comb filtering.
HONKY—The music sounds the way your voice sounds when you
cup your hands around your mouth. A bump in the response around
500 to 700 Hz.
IN YOUR FACE—Dry (without effects, without reverb), possibly
with compression.
LIQUID—Opposite of grainy. A sense of seamless flowing of the
music. Flat response and low distortion. High frequencies are flat or
reduced relative to mids and lows.
LOW-FI (low fidelity)—“Trashy” sounding. Tinny, distorted, noisy,
or muddy.
MELLOW—Reduced high frequencies, not edgy.
MUDDY—Not clear. Weak harmonics, smeared time response, distortion. Too much reverb at low frequencies. Too much emphasis
around 200 to 350 Hz. Too much leakage.
MUFFLED—The music sounds covered up. Weak highs or weak
upper mids.
MUSICAL—Conveying emotion. Flat response, low distortion, no
edginess.
NASAL—The vocalist sounds as if she is singing with the nose
closed. Also applies to strings. Bump in the response around 500 to
1000 Hz. See Honky.
NEUTRAL—Accurate tonal reproduction. No obvious colorations.
No serious peaks or dips in the frequency response.
PAPERY—Referring to a kick drum that has too much output around
400 to 600 Hz.
PHASEY—Having phase interference (comb filtering). The sound of
a direct signal and its delayed repetition mixed to the same channel (delay usually under 20 msec). Might be due to multiple mics
picking up the same source, or one mic picking up direct sound and
Effects and Signal Processors CHAPTER 10
delayed reflected sound. Or a delayed signal mixed with itself undelayed. Or, some opposite-polarity crosstalk between stereo channels.
Or one monitor speaker is reversed in polarity.
PIERCING—Strident, hard on the ears, screechy. Having sharp, narrow peaks in the response around 3 to 10 kHz.
PINCHED—Narrowband. Midrange or upper midrange peak in the
frequency response. Pinched dynamics are overly compressed.
PRESENT, PRESENCE—Adequate or emphasized response around
5 kHz for most instruments, or around 2 to 5 kHz for kick drum and
bass. Having some edge, punch, detail, closeness, and clarity.
PUFFY—Bump in the response around 400 to 700 Hz.
PUNCHY—Good reproduction of dynamics. Good transient
response. Or conversely, referring to highly compressed transients
(especially snare drum and kick drum) that sound like hitting a
punching bag. Sometimes a bump around 5 kHz or 200 Hz.
RASPY—Harsh, like a rasp. Peaks in the response around 6 kHz
which make vocals sound too sibilant or piercing.
RICH—See Full. Also, having euphonic distortion made of evenorder harmonics.
ROUND—High-frequency rolloff or dip. Not edgy.
SHARP—See Crisp, Strident, and Tight.
SIBILANT, ESSY—Exaggerated “s” and “sh” sounds in vocals, too
much output around 5 to 10 kHz.
SIZZLY—See Sibilant. Also, too much highs on cymbals.
SMEARED—Lacking detail. Poor transient response. This may be a
desirable effect in large-diameter mics. Also, poorly focused images.
SMOOTH—Easy on the ears, not harsh. Flat frequency response,
especially in the midrange. Lack of peaks and dips in the response.
Low distortion.
SPACIOUS—Conveying a sense of space, ambience, or room around
the instruments. To get this effect, mike farther back, mix in an ambience mic, add reverb, or record in stereo. Components that have
233
234
CHAPTER 10 Effects and Signal Processors
opposite-polarity or out-of-phase crosstalk between channels may
add false spaciousness.
SQUASHED—Overly compressed.
STEELY—Emphasized upper mids around 3 to 6 kHz. Peaky, non-flat
high-frequency response. See Glassy, Harsh, Edgy.
STRAINED—The component sounds like it’s working too hard.
Distorted. Inadequate headroom or insufficient power. Opposite of
effortless.
STRIDENT—See Harsh and Edgy.
SWEET—Not strident or piercing. Flat high-frequency response, low
distortion. Lack of peaks in the response. Highs are extended to 15
or 20 kHz, but they aren’t bumped up. Often used when referring to
cymbals, percussion, strings, and sibilant sounds.
THIN—Fundamentals are weak relative to harmonics. Note that the
fundamental frequencies of many instruments aren’t very low. For
example, violin fundamentals are around 200 to 1000 Hz. So if the
300 Hz area is weak, the violin may sound thin—even if the mic’s
response goes down to 40 Hz.
TIGHT—Good low-frequency transient response. Absence of ringing
or resonance when reproducing the kick drum or bass. Good lowfrequency detail. Absence of leakage.
TINNY, TELEPHONE-LIKE—Narrowband, weak lows, peaky mids.
The music sounds like it’s coming through a telephone or tin can.
TRANSPARENT—Easy to hear into the music, detailed, clear, not
muddy. Wide, flat frequency response, sharp time response, very low
distortion and noise.
TUBBY—See Bloated. Having low-frequency resonances as if you’re
singing in a bathtub.
VEILED—The music sounds like you put a silk veil over the speakers.
Slight noise or distortion, or slightly weak high frequencies.
WARM—Good bass, adequate low frequencies, adequate fundamentals relative to harmonics. Not thin. Or, excessive bass or midbass.
Or, pleasantly spacious, with adequate reverberation at low frequencies. Or, gentle highs, like from a tube amplifier. See Rich.
Effects and Signal Processors CHAPTER 10
WASHED OUT—Phase interference from multiple mics picking up
the same source. Too much leakage or reverberation.
WOOLY, BLANKETED—The music sounds like there’s a wool blanket over the speakers. Weak high frequencies or boomy low frequencies. Sometimes, an emphasis around 250 to 600 Hz.
235
This page intentionally left blank
CHAPTER 11
Mixers and Mixing
Consoles
The heart of your recording studio is the mixer. It’s a control center
where you plug in all sorts of signals; mix or blend them; add effects,
EQ, and stereo positioning; and route the signals to recorders and
monitor speakers. A mixing console (also called board or desk) is a
large mixer with many controls.
A recorder–mixer combines a mixer and a multitrack recorder in a
single portable chassis. This convenient unit is also called a ministudio, portable studio, or digital multitracker. Low-end recorder–
mixers record 4 tracks, and high-end units record 8 to 32 tracks.
STAGES OF RECORDING
First let’s review the three stages of using a mixer during a session:
recording, overdubbing, and mixdown.
1. Recording (tracking): The mixer accepts mic-level signals and
amplifies them up to line level. You send the line-level signal
from each mic to a separate track in a multitrack recorder. The
multitrack recorder records several tracks on hard disk. One
track might be a lead vocal, another track might be a saxophone, and so on.
2. Overdubbing: While listening to recorded tracks over headphones, the musician records new parts on open (unused)
tracks. You set up the mixer to monitor both the recorded
tracks and the new live part.
3. Mixdown: After all the tracks are recorded, you use the mixer
to combine or mix the tracks into 2-channel stereo with panning, EQ, and effects.
237
238
CHAPTER 11 Mixers and Mixing Consoles
MIXER FUNCTIONS AND FORMATS
Although the knobs, buttons, and meters on a mixer may appear
intimidating, you can understand them if you read the manual and
practice with the equipment. A mixer is complicated because it lets
you control many aspects of sound:
■
■
■
■
■
■
■
■
The loudness of each instrument (the balance among instruments in the mix)
The tone quality of each instrument (bass, treble, midrange)
The room that the instruments are in (reverberation)
The left-to-right position of each instrument (panning)
Effects (flanging, echo, chorus, etc.)
Track assignments (which instrument goes on which track)
Recording level (the signal voltage going to the recorder tracks)
Monitor selection (what you want to listen to)
Mixers come in many formats:
■
■
■
■
Analog mixer: A control device that works on analog signals,
and sends them to an external multitrack recorder and 2-track
recorder.
Digital mixer: The same as the analog mixer, but works internally with digital signals. It accepts analog or digital signals.
Software mixer: Exists only in your computer as part of digital
recording software. You control it with a mouse or a control
surface (described next). The recording is done on your hard
drive. Whether you use a hardware or software mixer, be sure to
read the section in this chapter on analog mixers because many
of the same principles apply.
Control surface: A device that looks like a mixer with faders and
knobs. It plugs into your computer’s USB or FireWire port and
controls the software mixer.
A mixer can be specified by the number of inputs and outputs it has.
For example, an 8-in, 2-out mixer (8 ⫻ 2 mixer) has 8 signal inputs
that can be mixed into 2 output channels (buses) for stereo recording. Similarly, a 16-in, 8-out (16 ⫻ 8) mixing board has 16 signal
inputs and 8 output channels for multitrack recording. A 16 ⫻ 4 ⫻ 2
mixing board has 16 inputs, 4 submixes or groups (explained under
the section Output Section) and 2 master outputs. There also are
connectors for external equipment, such as effects devices and a
Mixers and Mixing Consoles CHAPTER 11
monitor power amplifier. The more inputs your mixer has, the more
instruments you can record at the same time. If you’re recording only
yourself, you may need only two inputs.
Let’s look at the analog mixer in more detail. Knowing how it works
will help you understand the other types.
ANALOG MIXER
A mixer can be divided into three sections: input, output, and monitor. Here are the main parts of each section and what they do:
Input section
■
■
■
■
■
■
Inputs connect to your mics, electric instruments, and recorder
outputs.
Faders are sliding volume controls that affect the loudness of
each instrument. This lets you control the balance among
instruments in the mix.
EQ knobs adjust the tone quality of each instrument (bass, treble, midrange).
Aux knobs set the amount of reverb or other effects, and also
can be used to set up a monitor mix or headphone mix.
Pan pots place the monitored sound of each track where desired
between your stereo speakers—left, center, right, or anywhere
between. In some mixers, the pan pot is also used with the assign
switch during recording to send signals to the desired tracks.
Channel assign buttons route each input signal to the desired
recorder track.
Output section
■
■
■
■
Master faders set the overall level of the entire stereo mix.
Group faders set the overall level of each group or submix.
Outputs connect to your recorder inputs and the monitor power
amplifier.
Meters help you set the correct recording level (to prevent distortion and noise).
Monitor section
■
■
Monitor controls select what you want to listen to.
Aux knobs or channel faders set up the monitor mix.
Let’s look at each part in more detail.
239
240
CHAPTER 11 Mixers and Mixing Consoles
GAIN TRIM
aux 1
AUX
aux 2
HF
MF
EQ
LF
PAN
CLIP LED
FADER
ASSIGN
FIGURE 11.1
A typical input module.
Input Section
A mixer is made of groups of controls called modules. An input
module (Figure 11.1) or channel strip affects a single input signal—
from a microphone, for instance. The module is a narrow vertical
strip, one per input. Several modules are lined up side by side. Each
input module is the same, so if you know one, you know them all.
Let’s follow the signal flow from input to output through a typical input module (Figure 11.2). Every mixer is a little different, but
you’re likely to find features like those described here.
INPUT CONNECTORS
On the back of each input module are input connectors with these
labels:
MIC: Accepts signals from a microphone or direct box.
LINE: Accepts an electric musical instrument, or a track output of
a multitrack recorder.
TRACK: Accepts a track output of a multitrack recorder; not
included in all mixers.
Some units use a single jack for both mic and line inputs; others
have separate jacks for each. The mic input is either an unbalanced
Mixers and Mixing Consoles CHAPTER 11
GAIN
OR
TRIM
OVERLOAD
(CLIP) LED
INPUT
SELECTOR
SWITCH
MIC
INPUT
CONNECTORS
DIRECT
OUT
JACK
PREAMP
1
MUTE
EQ
PAN
POT
FADER
LINE
INSERT SEND
INSERT RETURN
CHANNEL
ASSIGN
SWITCH
PRE-FADER
AUX SEND
To the
3 group
mixing
2 circuits
4
POST-FADER
AUX SEND
FIGURE 11.2
Signal flow through a typical input module.
1/4-inch phone jack (a 1/4-inch hole) or an XLR-type connector
(with three small holes). The line input is either a 1/4-inch phone
jack, an RCA phono jack (like you see on a stereo system), or an
XLR-type connector.
You can plug a synthesizer directly into a phone-jack line input
without using a direct box if the cable is under 10 feet; a longer cable
may pick up hum. In that case, use a direct box plugged into a mic
input.
In some mixers, a phone-jack input is switchable between low
impedance (for microphones) or high impedance (for guitar pickups). Other mixers might have a separate low-impedance mic input
and a high-impedance instrument input.
PHANTOM POWER (P48, ⫹48)
This switch (not shown) turns on phantom power (48 V DC) for condenser microphones. In the mic input connector, the 48 V appears
on pins 2 and 3 relative to pin 1. The microphone receives phantom
power and sends audio along the same two cable leads.
MIC PREAMP
After entering the mic connector, the microphone signal goes to a
mic preamplifier inside the mixer. This preamp boosts or amplifies
241
242
CHAPTER 11 Mixers and Mixing Consoles
the weak microphone signal up to a higher voltage, making it a linelevel signal.
The TRIM or GAIN control adjusts the amount of amplification in
the mic preamp. If the gain control is turned up full, and the incoming mic signal is very strong due to a loud instrument or vocal, this
signal can overload the mic preamp. This causes distortion—a gritty
sound. To get a good signal-to-noise ratio (S/N) in your mixer, set
the gain control as high as possible, but not so high that the preamp
distorts.
Here’s how: Start with the trim turned up all the way down (counterclockwise). In most mixers, each input module has a tiny light
(LED) labeled “clip,” “peak,” or “OL” (overload). It flashes when the
mic preamp is distorting. When an instrument is playing its loudest
signal through an input module, gradually turn up the trim control
until the clip LED starts flashing. Then turn down the trim control
just to the point where the light stays off, and turn it down another
6 dB for extra headroom.
Some low-cost mixers don’t have a trim control. The input fader
serves this function.
INPUT SELECTOR SWITCH
This switch lets you select the type of signal you want to work with.
Some common switch labels are below:
MIC: (Mic or direct box)
LINE: (For line-level signals: a synthesizer, drum machine, electric
guitar, or multitrack recorder’s track output)
INPUT: (Mic or line)
TRACK: (Multitrack recorder’s track output)
Using the input selector is simple. If you plugged in a microphone or
direct box to record its signal, set the input selector to MIC or INPUT.
If you plugged in a synth, drum machine, or electric guitar, set the
selector to LINE or INPUT. When you’re ready to mix, select TRACK
(if available) or LINE (if the recorder track outputs are plugged into
mixer line inputs).
Some mixers have no input selector. The mixer processes whatever
signal is plugged in.
Mixers and Mixing Consoles CHAPTER 11
In some recorder–mixers, a single 1/4-inch phone jack is used both
for mic-level and line-level signals. The two levels are handled either
by a MIC/LINE switch or a TRIM control.
INSERT JACKS
Following the input selector switch are the insert-send jack and insertreturn jack (on the back of each input module). Or the mixer might
have a single insert jack, a TRS type that has the send on the tip terminal and the return on the ring terminal. Inside the mixer, the send
is connected to the return so that the signal passes through to the rest
of the mixer.
If you insert a plug into the insert jack(s), you break the signal path
so you can insert an external device there in series with the input
module’s signal. You might insert a compressor into the signal path
of one module for automatic volume control, or insert any other signal processor (reverb/delay, for instance). That way, if all your aux
sends are tied up, you can add another signal processor. On the
reverb/delay unit, set the dry/wet mix control for the desired amount
of effect.
Another use for the insert jack is to send the mic-preamp output signal to a multitrack recorder track. The output of each track returns to
the insert jack and continues through your mixer. In this case, you
use the trim controls to set recording levels, and use the faders, EQ,
and aux sends to set up a monitor mix.
Low-cost mixers omit the insert jacks. Some units have insert jacks
on only two inputs.
INPUT FADER (CHANNEL FADER)
Next, the input signal that you selected goes to a fader. This is a sliding volume control for each input signal. During recording, you can
use the fader in two ways. If you’re recording one instrument per
track, record off the insert send and use the fader to adjust the instrument’s level in the monitor mix. If you’re recording two or more
instruments on one track, record off a group output, and use each
instrument’s fader to set up a mix within that group. For example,
if you’re recording several drum mics onto one track, set up a drum
mix with the drum-mic faders.
243
244
CHAPTER 11 Mixers and Mixing Consoles
During mixdown, you use the faders to set the loudness balance
among instruments.
EQ
The signal from the input fader goes to an equalizer, which is a tone
control. With EQ you can make an instrument sound more or less
bassy, and more or less trebly, by boosting or cutting certain frequencies. Details are in Chapter 10 in the section Equalizer.
DIRECT OUT
The direct out is an output connector following each input fader and
equalizer. The signal at the direct-out jack is an amplified, equalized version of the input signal. The fader controls the level at the
direct-output jack. You can use the direct-out jack when you want to
record one instrument per track (with EQ) on an external multitrack
recorder. Connect the direct-out jack to a multitrack recorder track
input. Because the direct output bypasses the mixing circuits farther
down the chain, the result is a cleaner signal.
Suppose your mixer has 8 inputs and 2 outputs. You can use this
mixer with an 8-track recorder. Just connect the direct-out jack in
each input module to a separate track input. Or do the same with
the insert-send jacks.
CHANNEL ASSIGN SWITCH
The equalized signal also goes through a pan pot (explained in the
section below) to the channel assign switch or track selector switch.
It lets you send the signal of each instrument to the recorder track
you want to record that instrument on.
A mixer with four groups or buses would have an assign switch
labeled 1, 2, 3, 4. If you want to record bass on track 1, for instance,
find the assign switch for the input module the bass is plugged into,
and push assign switch 1. If you want to record four drum mics on
track 2, push assign switch 2 for all those input modules.
Some mixers assign tracks with two controls: a selector switch and a
pan pot.
Mixers and Mixing Consoles CHAPTER 11
PAN POT
This knob sends a signal to two channels in adjustable amounts.
By rotating the pan-pot knob, you control how much signal goes to
each channel. Set the knob all the way left and the signal goes to
one channel. Set it all the way right and the signal goes to the other
channel. Set it in the middle and the signal goes to both channels
equally.
Here’s how you might use a pan pot to assign an instrument to a
track. The channel-assign switch might have two positions labeled
1–2 and 3–4. If you turn the pan pot left, the signal goes to oddnumbered tracks (either 1 or 3, depending on how you set the
assign switch). If you turn the pan pot right, the signal goes to evennumbered tracks (2 or 4). Suppose you want to assign the bass to
track 1. Set the assign switch to 1–2, and turn the pan pot far left
to choose the odd-numbered track (track 1).
During mixdown, the pan pot has a different function: It places
images between your speakers. An image is an apparent source of
sound, a point between your speakers where you hear each instrument or vocal. Set the pan pot to locate each instrument at the left
speaker, right speaker, or anywhere in between. If you set the pan
pot to center, the signal goes equally to both channels, and you hear
an image in the center.
AUX
The aux or aux-send function (Figures 11.2 and 11.3) sends some of
the input module’s signal to equipment outside the mixer. A prefader aux send is before the fader, and usually goes to a power amp
that feeds monitor speakers or headphones. You can use the pre-fader
aux knobs to set up a monitor mix—a balanced blend of input signals you hear over speakers or headphones. A post-fader aux send is
after the fader and EQ, and usually goes to an effects unit. You use
the post-fader aux knobs to set the amount of effects (reverb, echo)
heard on each instrument in a mix.
Some mixers have no aux sends, some have one aux-send control per
module, and some have two or more (labeled aux 1, aux 2, etc.). The
more aux sends you have, the more you can play with effects, but
the greater the cost and complexity. The aux number (1 or 2) isn’t
245
246
CHAPTER 11 Mixers and Mixing Consoles
INPUT
FADER
GROUP
MIXING CIRCUIT
(TYPICAL OF 4)
EQ,
PAN,
ASSIGN
INPUT
SIGNAL
AUX 1
(EFFECTS)
Signals from
other input
modules
GROUP
OUT
AUX 1
SEND
Signals from
other aux 1
knobs
EXTERNAL
EFFECTS UNIT
AUX RETURN
OR BUS IN
AUX 2
(MONITOR)
Signals from
other aux 2
knobs
AUX 2
SEND
To monitor amp
FIGURE 11.3
Aux sends.
necessarily assigned a specific function; you decide what you want
aux 1 and aux 2 to do.
During recording and overdubbing, the aux knobs of all the input
modules can be used to create a monitor mix. The monitor mix that
you create with the aux knobs is independent of the levels going to
the multitrack recorder. You use the gain-trims during recording to
set recording levels, and you use the aux knobs to create an independent mix that is heard over your monitor system.
In Figure 11.3, the aux-2 send control is just before the fader. In
this mixer, the signals from all the aux-2 knobs in the mixer combine at a connector jack labeled “aux-2 send.” You can connect that
jack to your power amplifier, which drives monitor speakers and
headphones.
In Figure 11.3, the aux-1 send control is just after the fader. During
mixdown, each aux-1 knob controls how much effects (reverb, echo)
you hear on each track. In each input module, the aux knob adjusts
how much of that input signal is sent to an external effects unit.
The effected signal returns to the mixer’s aux-return or bus-in jacks,
where it blends with the original signal.
Mixers and Mixing Consoles CHAPTER 11
For example, suppose the aux-1 send is connected to a digital reverb.
The more you turn up the aux-1 send knob, the more signal goes to
the reverb. The output of the reverb unit returns to the mixer’s bus-in
jack, and blends with the original dry signal, adding a spacious effect.
A few mixers have an aux-return control (also called effects-return or
bus-in) that sets the overall effects level returning to the mixer.
There might be a pre/post switch next to the aux-send knob. When
an aux knob is set to pre (pre-fader), its level isn’t affected by the
fader setting. You use the pre setting for a headphone mix during
recording or overdubbing because you don’t want the fader settings
to affect the monitor mix.
The post setting (post-fader) is used for effects during mixdown. In
this case, the aux level follows the setting of the fader. The higher
you set the track volume with the fader, the higher the effects level.
But the dry/wet mix stays the same.
Output section
The output section is the final part of the signal path; the section that
feeds mixed signals to the recorder tracks. It includes mixing circuits,
submaster or group faders (sometimes), master faders, and meters
(Figure 11.4).
MIXING CIRCUITS (ACTIVE COMBINING NETWORKS), GROUP FADERS,
AND BUS OUTPUT CONNECTORS
The group mixing circuits are in the center of Figure 11.4. Recall that
you use the assign switches to send each input signal to the desired
channel or bus, and each bus feeds a different recorder track. A bus
is a channel in a mixer containing an independent mix of signals.
The bus 1 or group 1 mixing circuit accepts the signals from all the
inputs you assigned to bus 1 and mixes them together to feed track
1 of the recorder. The bus 2 mixing circuit mixes all the bus 2 assignments, and so on.
Following each group mixing circuit is a group fader. If you assigned
all the drum mics to group 1, the group 1 fader adjusts the overall
level of the drum mix. The signal from each group fader goes to a
group or bus output connector in your mixer. You can connect each
bus output to a recorder track input.
247
248
CHAPTER 11 Mixers and Mixing Consoles
OUTPUT SECTION
INPUT MODULE (one of several)
FADER
Input
signal
from mic
or line
EQ,
PAN
AUX SEND
To the
aux
mixing
circuit
Assign
switches
to the
group
mixing
circuits
GROUP
MIXING
CIRCUITS
METER
(TYPICAL OF 4)
1
Group 1 out
2
Group 2 out
Signals
from 1
2
group 3
outs 4
Signals from
the assign
switches of all
the input modules
4
Signals from
the aux sends
of all the
input modules
CH.1
STEREO
MIX BUS
GROUP
FADERS
3
MASTER
FADERS
Group 3 out
1
2
3
4
To recorder
Ch. 1 in
CH.2
To recorder
Ch. 2 in
Group 4 out
Aux-send mix
to monitor power amp
or effects device input
AUX
MIXING
CIRCUIT
FIGURE 11.4
Input module and output section of a mixer.
STEREO MIX BUS, MASTER FADERS, MAIN OUTPUT CONNECTORS
The stereo mix bus includes two group mixing circuits: one for channel 1 and one for channel 2. Three types of signals feed into the stereo mix bus:
1. The group output signals.
2. Signals from input modules. You can assign an input module’s
signal directly to the stereo mix bus, bypassing the groups. This
results in lower noise.
3. Effects-return signals, such as the reverberated signal from an
external digital reverb.
Located on the right side of your mixer, the stereo master faders
are one or two sliding volume controls that affect the overall level
of the stereo mix bus. Usually, you set the master fader(s) within
design center, the shaded area about three-quarters of the way up
on the scale. This setting minimizes mixer noise and distortion. You
can fade out the end of a mix by turning down the master faders
gradually.
After the master faders, the signal goes to a pair of main output connectors, which feed a 2-track recorder of your choice.
Mixers and Mixing Consoles CHAPTER 11
You feed the multitrack recorder either from group outputs, direct
outs, or insert sends. If you’re mixing several instruments to track 5,
for example, assign those instruments’ signals to Group 5. Connect
the Group 5 output to recorder track 5 in. If you’re recording one
instrument on track 5, however, connect that instrument’s direct out
or insert send to track 5 in. The signal is cleaner at the direct out or
insert send than at the group output.
Some mixing consoles have voltage controlled amplifier (VCA)
group faders. A VCA group fader acts like a remote control for groups
of channel faders. For example, you could assign each drum mic’s
fader to a group, which is an audio path, and also assign each drum
mic’s fader to a VCA group, which is a fader that controls all the
drum mic channels at once.
METERS
Meters are an important part of the output section. They measure the
voltage level of various signals. Usually, each group or bus output has
a meter to measure its signal level. If these buses feed the recorder
tracks, you use the meters to set the recording level for each track.
A mixer has either VU meters or LED bar graph meters.
A VU meter (now rarely used) is a voltmeter that shows approximately the relative loudness of various audio signals. Set the record
level so that the meter needle reaches ⫹3 VU maximum for most signals, and about ⫺6 VU maximum for drums, percussion, and piano.
That’s necessary because the VU meter responds too slowly to show
the true level of percussive sounds.
An LED bar graph meter has a column of lights (LEDs) that shows
peak recording level. Usually you set the recording level to peak near
⫺6 dB maximum.
If your console has VU meters, and it is feeding a digital recorder
with LED peak meters, set a 0 VU tone to read ⫺20 dB on the
recorder meters. This allows some headroom for peaks.
Monitor Section
The monitor section is used to control what you’re listening to. It
lets you select what you want to hear, and lets you create a mix over
249
250
CHAPTER 11 Mixers and Mixing Consoles
headphones or speakers to approximate the final product. The monitor mix has no effect on the levels going to your recorder.
During recording, you want to monitor a mix of the input signals.
During playback or mixdown, you want to hear a mix of the recorded
tracks. During overdubs, you want to hear a mix of the recorded
tracks and the instrument that you’re overdubbing. The monitor
section lets you do this.
MONITOR SELECT BUTTONS
These buttons let you choose what signal you want to monitor or
listen to. Because the configuration of these buttons varies widely
among different mixers, they aren’t shown in Figure 11.4.
If you want to use aux 2 as the monitor-mix bus, select the aux-2 bus
as the monitor source. That lets you listen just to the aux2 mix. Some
mixers have no monitor-select switches. Instead, you always monitor
the stereo mix bus.
MONITOR MIX CONTROLS AND CONNECTORS
One way to set up a monitor mix is with the aux knobs. Suppose you
want aux 2 to be the monitor mix. Connect the aux-2 send jack to
your power amp and speakers. Or if you’re using headphones, find
the monitor-select switch for the headphones, and set it to aux 2.
Turn up all the aux-2 knobs about halfway, then turn each knob up
or down to set a good loudness balance. You do this during recording or overdubbing.
Here’s another way to set up a monitor mix using the faders.
Connect your multitrack recorder ins and outs to the insert jacks
(Figure 11.5). Connect the insert-jack 1 tip (send) to track 1 in; connect track 1 out to the insert-jack ring (return). Make similar connections for the other tracks. Also connect the mixer’s monitor-out jacks
to your power amp and speakers. Monitor the stereo mix bus. With
this setup, use the trim controls to set recording levels. Use the faders
to set up a monitor mix, cue mix, or mixdown with EQ and effects.
It’s a convenient way to work.
During mixdown, monitor the stereo mix bus. This is automatic in
a DAW.
Mixers and Mixing Consoles CHAPTER 11
TRIM
(TO SET
RECORD LEVEL)
MULTITRACK
RECORDER
IN
OUT
FADER
SEND
SIGNAL
IN
PREAMP
RETURN
TO EQ
AND AUX
INSERT JACK(S)
FOR MONITOR MIX
FIGURE 11.5
Using insert jacks to send each input signal to a recorder track. The track signal returns to the mixer,
where you adjust level, panning, EQ, and effects.
SOLO
The SOLO button in an input module is actually a monitor function.
It lets you monitor one instrument or vocal at a time so you can hear
it better or so you can edit one track. By pressing two or more SOLO
buttons, you can monitor more than one input signal at a time.
Suppose you hear a buzz in the audio and suspect it may be in the
bass guitar signal. If you push the SOLO button in the bass guitar’s
input module, you’ll monitor only the bass guitar. Then you can easily hear whether the buzz is in that input.
On British consoles, the SOLO function is called pre-fader listen
(PFL) or after-fader listen (AFL).
MUTE
When you press the MUTE button in a channel, the signal in that
channel is turned off. During mixdown, it’s a good idea to mute tracks
that have nothing playing at the moment to reduce noise and leakage.
During overdubs, you might want to mute previous takes or certain
instruments to make a cleaner headphone mix for the musician.
Additional Features in Large Mixing Consoles
Large mixing consoles have more features than small mixers. If you’re
working only with a small mixer or a recorder–mixer, you might want
to skip this section.
251
252
CHAPTER 11 Mixers and Mixing Consoles
FOLDBACK (FB): Another name for cue or headphone mix.
PHASE (POLARITY INVERT): Used only with balanced lines,
this switch inverts the polarity of the input signal; that is, it
switches XLR pins 2 and 3 to flip the phase 180 degrees at all
frequencies. You might use it to correct a miswired mic cable
whose polarity is reversed. If you mic a snare drum top and
bottom, you need to invert the polarity of the bottom mic.
AUTOMATED MIXING CONTROLS: These controls (Read, Write,
Update, Record Automation, Play Automation) set up the console for various automation functions. With automation, a
memory circuit in the mixer remembers your console settings
and mix moves. After leaving a mix for several days, you can
recall the settings with the push of a button, then continue
working on the mix. More on automated mixing is at the end
of Chapter 12.
EFFECTS PANNING: This feature places the images of the effects
signals wherever desired between the monitor speakers. Some
consoles let you pan effects in the monitor mix as well as in
the final program mix.
EFFECTS RETURN TO CUE: This is an effects-return level control
that adjusts the amount of effects heard in the studio headphone mix. These monitored effects are independent of any
effects being recorded.
EFFECTS RETURN TO MONITOR: This effects-return control adjusts the amount of effects heard in the monitor mix.
These monitored effects are independent of any effects being
recorded.
BUS/MONITOR/CUE switch for effects return: A switch that feeds
the effects-return signal to your choice of three destinations:
program bus (for mixdown), monitor mix (control-room
speakers), cue mix (headphones), or any combination of the
three.
METER SWITCHES: In many consoles, the meters can measure
signal levels other than console output levels. Switches near
the meters can be set so that the meters indicate bus level, auxsend level, aux-return level, monitor-mix level, etc.
Those readings help you set optimum levels for the outboard devices receiving those signals. Too low a level results
in noise; too high a level causes distortion in the outboard
unit. For example, if the aux-return signal sounds garbled or
Mixers and Mixing Consoles CHAPTER 11
distorted, the cause may be an excessive aux-send level. Verify
that condition by checking the meters switched to read the aux
or effects bus.
DIM: A switch that reduces the monitor level by a preset amount
so you can talk (as in “dim the lights”).
TALKBACK: An intercom between the control room and studio. A
mic built into the console lets you talk to the musicians in the
studio when you push the talkback button.
SLATE: This function routes the control-room microphone signal
to all the buses so you can record the name of the tune and
take number.
OSCILLATOR or TONE GENERATOR: This is used to put a levelcalibration tone on a tape, or to reference the recorder’s meters
to those on the console. You also can use it to check signal
path, levels, and channel balance.
Some mixers include a FireWire or USB port. It sends the output
signal of each mic preamp to a separate track in your recording
software, and returns the stereo mix back to the mixer for monitoring. See Chapter 4 under the heading Computer Digital Audio
Workstation.
DIGITAL MIXER
So far we’ve looked at the analog mixing console, which works
entirely with analog signals. A digital mixing console accepts analog
or digital signals. It converts the analog input signals to digital, and
processes all signals internally in digital format. The signal stays in
the digital domain for all mixer processing. Level changes, EQ, and
so on are done by digital signal processing (DSP or computer calculations) rather than by analog circuits.
Some of the digital output signals from the mixer go through digitalto-analog (D/A) converters. The resulting analog signals feed a power
amp, effects unit, and so on.
Analog and digital consoles operate differently. To use EQ in an analog console, you find the channel you want to EQ, and adjust its EQ
knobs. To use EQ in a digital console, you press a button to select
the channel you want to equalize, and press an EQ button. Then the
EQ settings for that channel show up on an LCD screen. You press
253
254
CHAPTER 11 Mixers and Mixing Consoles
buttons and turn a knob to adjust the EQ frequency and boost/cut
for that channel.
Because one knob controls the EQ for all the channels, digital consoles have fewer controls than analog consoles. One knob or switch
can have several functions. This makes digital consoles a little
harder to operate because you can’t just reach for an EQ knob for
a particular channel. You have to press a few buttons to set the EQ
parameters.
Some models have an LCD touch screen that shows a bank of faders
with small EQ curves, compression curves, and virtual aux knobs. To
tweak EQ for one channel, find the EQ curve and tap it. The screen
zooms to a larger view of the EQ settings for that channel. Rotary
encoders (knobs) under the touch screen, or built into it, let you
adjust parameters.
Digital consoles vary a lot in how they implement functions, so you
need to read the manual.
Some digital mixers have a bank switch or layer switch. It lets you
select the channels that the faders control. For example, suppose
a mixer has 16 faders. Press the layer switch once to make the faders access channels 1–16. Press the switch again to make the faders
access channels 17–32, and so on.
Several digital consoles have built-in effects, multitrack recording to
hard disk, and a USB port for a Flash drive that saves presets and
mixes.
Many digital mixers offer automated mixing. One type of automated
mixing is called scene or snapshot automation. When you press the
snapshot button, a memory circuit in the mixer takes a “snapshot”
of all the mixer settings for later recall. Another type of automation
is dynamic. Your mix moves are stored in memory for later recall.
When you recall a mix, some mixers make the faders move into the
positions you set up. This feature is called “flying faders” or “motorized faders.” A few mixers don’t move the faders when you recall a
mix. You have to set them manually by looking at a display, which is a
disadvantage. Motorized faders are more expensive but easier to use.
Mixers and Mixing Consoles CHAPTER 11
Digital Mixer Features
Look for the following features in digital mixers when making a buying decision:
■
■
■
■
■
■
■
■
■
■
■
■
■
Number of mic inputs and number of output buses
Number and type of digital inputs and outputs: S/PDIF, AES3,
MADI, TDIF, Lightpipe, USB, and FireWire. Chapter 9 describes
those standards under the heading Digital Audio Signal
Formats.
Word-clock input and output
Routing section that lets you route any input signal to any channel, or route any output signal to any output bus
Number and type of option card slots: extra I/O, DSP, sync,
effects, plug-ins
Number of effects processors
PC USB interface for control and/or backup
Ease of use
Snapshot or dynamic automation, or no automation
Motorized or non-motorized faders
Varispeed clock for pitch shift
Surround-sound monitoring
Touch screen with virtual controls
SOFTWARE MIXER
Here’s another type of mixer. A software mixer (virtual mixer) is a
simulated mixer you see on a computer screen (Figure 11.6). It exists
only in your computer as part of digital recording software or a DAW.
You control it with a mouse or with a controller surface. Recordings
made with this mixer, and all the mixer settings, are stored on your
hard drive.
In a software mixer, the input and output connectors are in an audio
interface (such as a sound card) that connects to the computer. This
interface is 2-channel or multichannel. The software mixer works
with effects that are either plug-ins (software) or external (hardware).
Automation is a standard feature. For more on software mixers, see
Chapter 13.
255
256
CHAPTER 11 Mixers and Mixing Consoles
FIGURE 11.6
Cakewalk Sonar Producer mix window, an example of a software mixer.
CONTROL SURFACE
A control surface or controller resembles a hardware mixer with faders and knobs (Figure 13.10). It plugs into your computer’s USB or
FireWire port and controls the software mixer. Some people prefer a
controller because it is easier to use than a mouse. More in Chapter 13
under the heading Control Surface.
Now that you understand the typical features of mixers and mixing
consoles, you’re ready to learn how to use them.
CHAPTER 12
Mixer Operation
Get your hands on those knobs. You’re going to operate a mixer as
part of a recording session. This will be a basic run-through. Detailed
session procedures are described in Chapter 15. Most of these procedures also apply to operating a virtual mixer in a Digital Audio
Workstation (DAW).
First recall the stages in making a recording:
1.
2.
3.
4.
5.
Session preparation
Recording
Punching in
Overdubbing
Mixdown
This chapter considers each stage in turn.
SESSION PREPARATION
If you’re recording on a hard drive, make sure you have enough drive
space for the project. Chapter 13 has simple equations and a table
that show how much space you need.
Plan your track assignments. Write a track sheet that tells what instrument goes on which track. Note: If you assign multiple instruments
to the same track, you can’t separate their images in the stereo stage.
That is, you can’t pan them to different positions; all the instruments
on one track sound as if they’re occupying the same point in space.
If you’re recording 4-track, you may want to do a stereo mix of the
rhythm section on tracks 1 and 2; then overdub vocals and solos on
tracks 3 and 4.
257
258
CHAPTER 12 Mixer Operation
Studio setup for the musicians is covered in Chapter 15.
Set Up the Mixer and Recorder
To start the process, first zero or neutralize the mixer by setting all
the controls to “off,” “flat,” or “zero.” This establishes a point of reference and avoids surprises later on. Set all faders down.
If you have a separate mixer and multitrack recorder, you need to
make their meter readings match. Follow one of these procedures:
If your mixer and recorder have VU meters, play a steady tone into
the mixer to get a 0 reading on the meters for all channels. Then set
the multitrack recorder’s record level (if any) to get 0 VU readings on
all the tracks.
If your mixer has VU meters and your recorder has LED peakreading meters, set 0 VU on your mixer to equal ⫺20 dB on the
recorder’s peak meters.
If your mixer and recorder have LED peak-reading meters, set 0 dB
on the mixer to equal 0 dB on the recorder’s peak meters.
If you’re feeding the multitrack recorder from your mixer’s insert
jacks or direct-out jacks, you’ll need to watch the recorder meters.
Suppose you’re ready to record a vocal or an acoustic instrument.
Place the microphone and plug it into a mic input. If you want to
record the audio output of a hardware synth or drum machine, connect a cable between the instrument’s output and a line input on the
mixer.
Attach a “scribble strip” of masking tape or white removable tape
along the bottom of the faders, and label each fader according to
what instrument you plugged into that input. Some consoles and
software mixers have scribble strips that you can type on. In Figure
12.1, such a strip is on the mixer at the right.
Set the input selector (if any) to Mic or Line depending on what is
plugged into each input.
Plug in headphones to hear what you’re recording. Turn up the
headphone volume control. Or, if you’re in a control room and the
musicians are in a studio, turn up the monitor level to listen over
the monitor loudspeakers. On a hardware mixer, set the MONITOR
Mixer Operation CHAPTER 12
FIGURE 12.1
A studio control room with a DAW, mixer and monitor speakers.
SELECT switch to hear the signal you’re recording, and turn up the
musicians’ cue mix.
Set the master faders about three-quarters of the way up, at 0, or
within the shaded portion of fader travel. This is called design center.
If you’re recording a soft synth—a software synthesizer in a DAW—
there will be a short delay in the monitored signal because of the
latency (signal processing delay) in the software. Set the latency or
buffer setting as low as possible without getting drop-outs. A delay
under 5 msec or so should be acceptable.
If you’re recording a vocal or real instrument with a DAW, don’t
enable Echo input monitor because this causes a latency delay in the
monitored signal. Instead, monitor the mic or input signal directly
from your mixer or audio interface.
259
260
CHAPTER 12 Mixer Operation
Assign Inputs to Tracks
If your equipment wiring is mics ⬎ mixer ⬎ insert sends ⬎ track
inputs, your inputs are already assigned. Input 1 goes to track 1;
input 2 goes to track 2, and so on.
If you’re using a computer DAW, select the input signal source for
each track. For example, you might set track 5’s input to channel 5 in
your audio interface. Then, whatever is plugged into interface channel 5 will go to track 5. If your sound card has just 2 channels, set
track 5’s input to channel 1 or 2 in your audio interface, and likewise
for the other tracks. Some DAWs record stereo tracks, so you’d assign
each track’s input to stereo channels 1 and 2 in the audio interface.
If your equipment wiring is mics ⬎ mixer ⬎ buses ⬎ track inputs,
assign each input signal to the desired output channel (bus) as specified on your track sheet. Each bus is connected to the corresponding numbered track on the multitrack recorder. Note: If only one
instrument is assigned to a track, you can eliminate the noise of the
console’s combining amplifier by patching the instrument’s signal
directly to the recorder track. To do that, locate the direct output
jack (or the insert jack) of the input module for that instrument and
patch it to the desired track. Some mixing boards also require pressing a Direct button on the input module.
Set Recording Levels
Now you’re ready to “get a level.” Have each instrument play the
loudest part of the music, one at a time or all at once. For each input
signal, set the TRIM control so the recording level is as high as possible without causing distortion. While setting levels for a digital multitracker, peak each track around ⫺6 dBFS maximum in peak meter
mode, not rms. This allows some headroom for surprises. Also,
musicians generally play louder during a performance than during
a level check. If you exceed 0 dBFS you’ll hear digital clipping which
makes a loud click.
Some audio interfaces use on-screen volume controls to set recording levels. Other interfaces have level knobs for this purpose.
If you are mixing several instruments to one or two groups (as in a
drum submix), follow this procedure:
1. Monitor the group(s).
2. Set its submaster fader (group fader) to design center.
Mixer Operation CHAPTER 12
3. Set the submix balances, panning, and recording level with the
input faders.
4. Fine-tune each submix level with the submaster fader.
Set EQ
Although it’s common to record flat (without EQ), you may want
to apply equalization at this point to each instrument heard individually. Filter out frequencies above and below the range of the
instrument. However, don’t spend too much time on EQ until all
the instruments are mixed together. The EQ that sounds right on a
soloed instrument may not sound right when all the instruments
are heard together. In creating the desired tonal balance, use EQ as
a last resort after trying different mics and mic placements. You also
can apply EQ during playback or mixdown. This may be preferable
because EQ applied when recording can’t always be undone if you’re
unhappy with it.
RECORDING
Next, set the track(s) you want to record to “record ready” mode.
Now start recording. Write down the counter time for this take.
“Slate” the recording: record the name of the tune and the take number. Then record a two-measure count-off. This is done to set the
tempo for overdubs. For example, if the time signature is 4/4, you
say, “1, 2, 3, 4, 1, 2, (rest) (rest).” The rests are silent beats. You need
some silence before the song starts to make editing easier later on.
A DAW has a built-in metronome (click track) that can be used for
count-offs.
PLAYBACK
After recording the track, go to the beginning of the song using the
return-to-zero or locate function. If necessary, set the monitor selector to “track,” or “mix,” and play back the recording to check the
performance and sound quality. You can set a rough mix with the
monitor mix (aux) knobs. If you’re recording on a computer, or if
your multitrack is patched to the insert jacks, use the faders, pan
pots, EQ, and aux knobs to set a rough mix.
Get or write a sheet of the song’s arrangement. It shows the lyrics,
verses, choruses, bridge, and so on. Play the song and set a locate
261
262
CHAPTER 12 Mixer Operation
point (marker) at the start of each verse and chorus. That way, when
the musician says “Let’s fix that flat note at the end of the second
chorus,” you can go instantly to that part of the song.
OVERDUBBING
After your first track is recorded with a good performance, you might
want to add more musical parts. This procedure is called overdubbing. When you overdub, the musician listens to tracks already
recorded, and records a new part on an unused track.
Ready to overdub? Here is one way:
1. Turn off the control room speakers and listen on headphones.
2. Set up a headphone monitor mix of the recorded tracks. Some
mixers use the aux knobs for this. If you’re recording on a computer, or if your multitrack recorder is connected to the mixer’s
insert jacks, monitor the stereo mix, and set up a mix with the
channel faders.
3. Plug in the mic or direct box for the instrument or vocal
you want to record. Assign it to an unused track. Turn up its
fader to design center (the 0 point about three-quarters of the
way up).
4. Set up your multitrack recorder, mixer, or audio interface to
monitor the playback of recorded tracks and the input signal
of the instrument or vocal you want to record.
5. Have the musician play or sing. Can you hear the signal in the
phones? Set the recording level using the TRIM (gain) control
for that mic’s channel.
6. Play the recording. As the musician plays or sings along,
set up a good mix of the recorded tracks and the live instrument you’re going to record. Ask the musician if he is hearing
what he needs to hear. Change the mix if needed. Some musicians want to hear effects when they overdub; some want it
dry. If you’re overdubbing backup vocals one at a time, often
it helps to remove (mute) certain other vocals from the headphone mix.
7. When you’re ready to record the new part, go to a point about
10 seconds before the part of the song where the musician plays.
Mixer Operation CHAPTER 12
Set the recorded tracks to SAFE and set the track you’re recording
on to RECORD READY.
8. Before you hit that RECORD button, stop! Are you recording
on the correct track(s)? Are you going to accidentally erase
any tracks? Double-check your track sheet, and make sure the
record-enable buttons are on only for the tracks that are safe to
record over.
9. Start recording and have the musician play along with the tracks. If
the musician makes a mistake, you can re-record or punch-in the
new part without affecting the parts that were already recorded.
PUNCHING-IN
Punching-in is used to fix mistakes in a recorded performance, or to
record a musical part in segments. A punch is also called an insert.
You enable record mode on a track, play the multitrack recording,
then “punch” or press the record button at the right spot, record a
new part, then punch-out of record mode.
Some musicians record the same musical part over and over until the
part is perfect. This process is tedious, but you have to pay attention.
You need to be aware of where you are in the song at all times, and
not erase anything you want to keep.
Some musicians like to record a performance a phrase at a time, perfecting each phrase as they go. Others record a complete take, then
go back and fix the weak parts.
To do a punch-in, grab your song-arrangement sheet and follow
these steps:
1. Go to a point about 10 seconds before the point where you want
to start recording. Note the counter time and write it down.
Also enter a memory location point there if your recorder can
do that.
2. Play the song to the musician over headphones. The musician
plays along to practice the part. Write down the counter times
where you want to punch-in and punch-out. Otherwise you
might erase a good take.
3. Finally you’re ready to record. Before you hit that RECORD
button, stop! Tell the musician what you’re going to do so
263
264
CHAPTER 12 Mixer Operation
there’s no chance of a mistake. Be very clear. For example, “I’m
recording your keyboard part on two new tracks.” Or, “I’m
punching-in over your old performance—is that what you
wanted?”
4. Okay, ready to go. Play the recording. During a rest or pause in
the music just before the part needing correction, punch-in the
record button (or use a footswitch). Have the musician record
the new part, and punch-out right away. You don’t want to
erase the rest of the track! If you’re using a DAW, though, you
can undo the recording.
5. Press LOCATE or GO TO MARKER to go to the location point
you set, about 10 seconds before the punch. Play the recording
to see if the punch was okay. If necessary, you can re-record
the punch. After you go to the locate point, notice whether
the musician wants to practice the part. Don’t redo the punch
until he is ready.
Some multitrack recorders have an autopunch function. You set
the punch-in and punch-out points into the machine’s memory. As
the recording plays, the recorder automatically goes into and out of
record mode at those points. Some recorders can loop repeatedly
between those two points.
COMPOSITE TRACKS
If several open tracks are available, you can record a solo performance in several takes, each on a separate track or virtual track. Then
combine the best parts of each track into a single track. Use only that
track in the final mix, and you’ll hear the best parts of all the takes
in succession. This is called “recording composite tracks” or “comping.” Here’s an example of comping vocal tracks with a multitrack
recorder and mixer:
1. Suppose you have four takes of a vocal recorded on tracks 10,
11, 12, and 13. Solo each track, and mark on the lyric sheet
which tracks have the best performance of each section. For
example, you might prefer track 13 on Verse 1, track 10 on
Chorus 2, and so on.
2. Assign all the vocal tracks to an open track, which we’ll call the
comp track. Match their levels.
3. Start recording on the comp track.
Mixer Operation CHAPTER 12
4. As the song plays, mute and unmute the tracks to copy the
best performances to the comp track. To do that, you can stop
the recorder between sections, change the mute settings, and
record a section at a time.
5. Once you’re happy with the comp track, you can erase or
archive the original tracks.
Comping with a DAW is even easier. Pick the best overall track, then
copy and paste good sections from other tracks into the best track.
Or divide the tracks into short sections, delete sections you don’t
like, and play the mix of all the tracks.
Cakewalk Sonar Producer software lets you record several takes on
one track, view all the take waveforms as “lanes” in that track, select
the best parts of each take, and create a composite track of those
parts.
Some recorders have virtual tracks (explained in Chapter 9 in the
section Recorder–Mixer (Digital Personal Studio). They let you comp
a performance with virtual rather than real tracks.
GETTING MORE TRACKS
Most DAWS can record nearly an unlimited number of tracks. But
what if you want to overdub more parts on a recorder–mixer, and
all the tracks are full? With care, you can punch-in more instruments
by recording them in the pauses on recorded tracks. For example,
suppose all the tracks are full but you want to add a cymbal crash
at the beginning of the chorus. Find a track that has a pause at that
moment, and punch in the cymbal crash there.
As an alternative, you can bounce tracks—mix several tracks to one
or two open tracks, and record the mix on that track. Then you can
erase the original tracks, freeing them for more overdubs. Bouncing
procedures are in the instruction manual for your recorder or
software.
DRUM REPLACEMENT
Suppose you’ve recorded the drum tracks, but you don’t like the
sound of them. Maybe the kick drum is too flabby, and no amount
of EQ or gating seems to help. You might try a technique called
265
266
CHAPTER 12 Mixer Operation
drum replacement. In a MIDI/audio recording program, you can
replace an audio drum track with a MIDI track that plays drum samples. First, set up a MIDI track with a drum sample ready to play (for
details, see Chapter 16). Then use one of these methods:
Method 1: In the audio drum track, insert a drum-replacement
plug-in such as Drumagog or Replacer. Set it to trigger the MIDI
track.
Method 2: Start recording on the MIDI drum track, and tap your
MIDI controller key in sync with the playback of the audio
drum track.
Method 3: Select the audio drum track, then enable the Extract
Rhythm or Extract Beat feature, if any. Copy and paste the
extracted MIDI beats to the MIDI drum-sample track.
Mute the original audio drum track and press PLAY. You should hear
the replacement drum sample playing.
MIXDOWN
After all your tracks are recorded (maybe with some bouncing), it’s
time to mix or combine them to 2-track stereo. You will use the
mixer faders to control the relative volumes of the instruments, use
panning to set their stereo position, use EQ to adjust their tone quality, and use the aux knobs to control effects. Figure 12.2 shows the
controls you will be using.
Set Up the Mixer and Recorders
Set all the mixer controls to “off,” “zero,” or “flat.” You should start
from ground zero in building a mix. Some engineers like to set all
the channel faders to ⫺12 dB and go from there.
Tape a strip of paper or masking tape along the front of the mixer
to write which instrument(s) each fader affects. Keep this strip for
use each time you play the recording. Or type in this information in
your DAW or mixing console.
If necessary, set the input-selector switches on the mixer to “track”
because you’ll be mixing down the multitrack recording. Monitor the
2-track stereo mix bus. These steps are unnecessary if you’re using
a DAW.
Mixer Operation CHAPTER 12
GAIN TRIM
aux 1
AUX
aux 2
HF
MF
EQ
LF
PAN
CLIP LED
FADER
ASSIGN
FIGURE 12.2
One input channel in a mixer, showing the fader, pan pot, EQ controls, and aux knobs.
For starters, put the master faders at design center (about threequarters of the way up, at the shaded portion of fader travel). This
sets the mixer gain structure for the best compromise between noise
and headroom.
Erase or Delete Unwanted Material
Mixing will be a lot easier if you first erase noises and leakage before
and after each song, and within each track.
Play the multitrack recording and listen to each track alone. Erase or
delete unwanted sounds, silent (unplayed) sections, outtakes, and
entire segments that don’t add to the song. In a DAW you can do this
visually by looking for parts of each track where there is no waveform. Zoom into a single track, highlight a silent portion, and select
Edit ⬎ Cut (or a similar command). To avoid mistakes, it’s best to
do this while the musicians are around. If you aren’t sure whether
it’s okay to delete part of a track, set that part’s volume envelope to
zero or mute the track during that part.
You can use editing to improve a song’s arrangement. Typically,
you’d put instrumental solos and fills only in the “holes” where the
vocal is silent. Add instruments to the mix gradually so that it builds
267
268
CHAPTER 12 Mixer Operation
in excitement. You might add vocal harmonies only in the choruses.
Reserve “all-out improvs” (where everybody plays) for the end.
Panning
You need to pan the tracks before doing the mix, because the loudness of a track depends on where it’s panned. Assign each track to
buses 1 and 2 (or the stereo mix bus), and use the pan pots to place
each track where desired between your stereo speakers. Typically
the bass, snare, kick drum, and vocals go to center; guitars can be
panned left or right, and stereo keyboards and drum overheads go
partly left and right (see Figure 14.2). Sometimes a mono drum mix
can be punchier than a stereo one.
You can give an acoustic guitar a spacious effect by recording it twice.
Pan the first take hard left and pan the second hard right. The second
take could be capoed higher to add interest.
Pan tracks to many points between the monitors: left, half-left, center, half-right, right. Try to achieve a stereo stage that is well balanced
either side of center. For clarity, pan to opposite sides any instruments that cover the same frequency range.
You may want some tracks to be unlocalized. Harmony singers and
strings should be spread out rather than appearing as point sources.
Stereo keyboard sounds can wander between speakers. You could
fatten a lead-guitar solo by panning it left, and panning the solo
delayed to the right. (This delay might come from a distant room
mic you used while recording, or from copying and sliding a track a
few milliseconds in a DAW.) Pan doubled vocals left and right for a
spacious effect.
Consider creating some front-to-back depth. Leave some instruments
dry so they sound close; add reverb to others so they sound farther
away.
If you want the stereo imaging to be realistic (say, for a jazz combo),
then pan the instruments to simulate a band as viewed from the
audience. If you’re sitting in an audience listening to a jazz quartet, you might hear drums on the left, piano on the right, bass in
the middle, and sax slightly right. The drums and piano aren’t point
sources, but are somewhat spread out. If spatial realism is the goal,
Mixer Operation CHAPTER 12
you should hear the same ensemble layout between your speakers.
In some rock recordings, the piano and drums are spread all the way
between speakers—interesting but unrealistic.
Pan-potted mono tracks often sound artificial; each instrument
sounds isolated in its own little space. It helps to add some stereo
reverb. It surrounds the instruments and “glues” them together.
When you monitor the mix in mono, you’ll likely hear center
channel buildup. Instruments in the center of the stereo stage will
sound louder in mono than they did in stereo, so the mix balance
will change in mono. To prevent this, note which tracks are panned
hard left or right, and bring them a little toward the center: 9 and
3 o’clock on the pan knobs.
Highpass Filter Each Track
To reduce muddy buildup, highpass all tracks by ear. Set up a highpass (low-cut) filter in each track. Adjust the filter frequency: start at
a low frequency, then gradually raise it until the sound starts to thin
out. Then back off a little (lower the filter frequency slightly).
Compression
Sometimes the lead vocal track might be too loud or too quiet relative to the instruments because vocals have a wider dynamic range
than instruments. You can control this by running the vocal track
through a compressor. It will keep the loudness of the vocal more
constant, making it easier to hear throughout the mix. If you’re using
a hardware mixer, patch the compressor’s inputs and outputs into
the insert jack(s) of the vocal input module. Or connect the compressor between the recorder’s vocal-track output and the mixer
input. Set the desired amount of compression ratio, attack time,
release time, and gain reduction. It’s also common to compress the
kick drum and bass. (For more information, see Chapter 10 under
the heading Compressor.)
Set a Balance
Now comes the fun part. The mixdown is one of the most creative parts
of recording. Here are some tips to help your mixes sound terrific.
269
270
CHAPTER 12 Mixer Operation
Before doing a mix, tune up your ears. Play over your monitors some
CDs whose sound you admire. This helps you get used to a commercial balance of the highs, mids, and lows.
Choose a CD with tunes like those you’re recording. Check out the
production. How is the balance set? How about EQ, effects, and
sonic surprises? Try to figure out what techniques were used to create those sounds, and duplicate them. Of course, you might prefer to
break new ground.
Using the input faders, adjust the volume of each track for a pleasing
balance among instruments and vocals. You should be able to hear
each instrument clearly. Some mixing consoles have trim knobs that
set the playback gain of the multitrack recorder tracks. In that case,
set all faders in use to design center, and adjust the trims to get a
rough mix.
Recording engineer Roger Nichols offers this suggestion: Change the
level of one track by ⫹1 dB, then by ⫺1 dB. Is it better? Is it worse?
If you can hear less than 1 dB change, the fader is close to where it
should be.
Here’s one way to build the mix. Make all the instruments and vocals
equally loud. Then turn up the most important tracks and turn down
background instruments. Or, bring up one track at a time and blend
it with the other tracks. For example, first bring up the kick drum to
about ⫺10dB, then add bass and balance the two together. Next add
drums and set a balance. Then add guitars, keyboards, and finally
vocals.
In a ballad, the lead vocal is usually on top. You might set the soloed
lead-vocal level to peak at ⫺6 dB. Bring up the monitor level so that
the vocal is as loud as you like to hear it, then leave the monitor level
alone. Bring in the other tracks one at a time and mix them relative
to the vocal track.
When the mix is right, everything can be heard clearly, yet nothing
sticks out too much. The most important instruments or voices are
loudest; less important parts are in the background. In a typical rock
mix, the snare is loudest, and the kick is nearly as loud. The lead
vocal is next in level. Note that there’s a wide latitude for musical
interpretation and personal taste in making a mix.
Mixer Operation CHAPTER 12
Sometimes you don’t want everything to be clearly heard. Once in a
while, you might mix in certain tracks very subtly for a subconscious
effect.
Recording engineer Michael Cooper offers this advice: To check the
vocal level, turn the monitors way down. The lead vocal should disappear just after the instruments do. To avoid washy sound, mike
acoustic guitars in mono, keep most tracks dry, use delay instead of
reverb, or use pre-delay with short reverbs.
Engineer Michael Stavrou (who wrote Mixing With Your Mind) suggests: Turn up the most inspiring player. Mix other tracks with it, one
at a time, to see which tracks fight it and which don’t. Mute or delete
the tracks that fight it.
If you recorded some MIDI tracks, consider replacing a MIDI instrument with another one that sounds better. You might replace a grand
piano with a honky-tonk piano, replace a fretted bass with a fretless
bass, and so on. Swap out the snare sound with a new one during
choruses.
It’s a good idea to monitor around 85 dBSPL. If you monitor louder,
the bass and treble will be weak when the mix is played softly.
To test your mix, occasionally play the monitors very quietly and see
if you can hear everything. Switch from large monitors to small, and
make sure nothing is missing.
Set EQ
Next, set EQ for the tonal balance you want on each track. If a
track sounds too dull, turn up the highs or add an enhancer. If a
track sounds too bassy, turn down the lows, and so on. Cymbals
should sound crisp and distinct, but not sizzly or harsh; kick drum
and bass should sound deep, but not overwhelming or muddy.
Be sure the bass is recorded with enough edge or harmonics to be
audible on small speakers.
For suggestions on EQ settings, see Chapter 10 under the heading
How to Use EQ.
You’ll need to readjust the mix balances after adding EQ. The EQ
that sounds right on a soloed track seldom sounds right when all the
tracks are mixed together. So make EQ decisions when you have the
271
272
CHAPTER 12 Mixer Operation
complete mix happening. Switch EQ in and out to make sure it’s an
improvement.
In pop-music recordings, the tone quality or timbre of instruments
doesn’t have to be natural. Still, many listeners want to hear a realistic timbre from acoustic instruments, such as the guitar, flute, sax, or
piano.
The overall tonal balance of the mix shouldn’t be bassy or trebly;
that is, the perceived spectrum shouldn’t emphasize lows or highs.
You should hear the low bass, mid-bass, midrange, upper midrange,
and highs roughly in equal proportions. Frequency bands that are
too loud can tire your ears.
When your mix is almost done, switch between your mix and a commercial CD to see whether you’re competitive. If the tonal balance of
your mix matches a commercial CD, you know your mix will translate to the real world. This works regardless of what monitors you
use. An effective tool for this purpose is Harmonic Balancer (www
.har-bal.com).
Add Effects
With the balances and EQ roughed in, it’s time to add effects. If
you’re using a DAW, use the plug-ins. If you’re using a hardware
mixer, you might want to plug in a reverb or delay to add spaciousness to the sound (see Chapter 10). This device connects between
your mixer’s aux-send and aux-return jacks (or aux-send and bus-in
jacks).
1. On the effects unit, set the dry/wet mix control all the way to
wet or effect.
2. Find the AUX RETURN or BUS IN controls (if any), set them
half of the way up, and pan them hard left and right.
3. Turn up the aux-send knob for each input, according to how
much effect you want to hear on that input signal. Suppose
you’re using reverb as an effect. You might turn it up by different amounts for the vocals, drums, and lead guitar, and leave it
turned down for the bass and kick drum.
4. As you’re setting the aux levels, check the overload indicator
on the effects unit. If it’s flashing, turn down the input level
on the effects unit just to the point where the overload light
Mixer Operation CHAPTER 12
stops flashing. Then turn up the output level on the effects unit
(or turn up the aux return on the mixer) to achieve the same
amount of effect you heard before.
Too much effects and reverb can muddy the mix. You might turn up
the reverb only on a few instruments or vocals. Once you have the
reverb set, try turning it down gradually and see how little you can
get by with. Typically the bass gets no reverb so it retains its clarity.
Usually, short reverb decay times (under about 1 second) are best for
up-tempo songs. Longer reverb times are better for slow ballads.
You might use 45 to 120 msec of pre-delay on vocals or certain
instruments to separate the dry sound from the reverb. This makes
the voice sound up-front even with reverb added.
The producer of a recording is the musical director and decides how
the mix should sound. The producer might be the band members or
yourself. Ask to hear recordings having the kind of sounds the producer desires. Try to figure out what techniques were used to create
those sounds.
Also try to translate the producer’s sound-quality descriptions into
control settings. If the producer asks for a “warmer” sound on a
particular instrument, turn up the low frequencies. If the lead guitar needs to be “fatter,” try a stereo chorus on the guitar track.
If the producer wants the vocal to be more “spacious,” try adding
reverb, and so on. Audio clip 39 at www.esevierdirect.com/companions/
9780240811444 demonstrates a mixdown.
Set Levels
Set the overall recording level as you’re mixing. To maintain the correct gain staging, keep the master faders at design center. Then adjust
all the input faders by the same amount so your stereo output level
peaks around ⫺6 dB. You can touch up the master faders a few decibels if necessary. Don’t exceed a 0 dB recording level if you’re recording to a digital medium.
Judging the Mix
When you mix, your attention scans the inputs. Listen briefly to each
instrument in turn and to the mix as a whole. If you hear something
273
274
CHAPTER 12 Mixer Operation
you don’t like, fix it. Is the vocal too tubby? Roll off the bass on the
vocal track. Is the kick drum too quiet? Turn it up. Is the lead-guitar
solo too dead? Turn up its effects send.
Check the mix while listening from another room, where the lows
and highs are weakened. Is the balance still good?
The mix must be appropriate for the style of music. For example, a
mix that’s right for rock music usually won’t work for folk music or
acoustic jazz. Rock mixes typically have lots of production EQ, compression, and effects; and the drums are way up front. In contrast,
folk or acoustic jazz is usually mixed with no effects other than slight
reverb, and the instruments and vocals sound natural. A rock guitar
typically sounds bright and distorted; a straight-ahead jazz guitar
usually sounds mellow and clean.
Suppose you are mixing a pop song, and you’re aiming for a realistic, natural sound. Listen to the reproduced instruments and try
to make them sound as if they’re really playing in front of you.
That is, instead of trying to make a pleasant mix or a sonically interesting recording, try to control the sound you hear to simulate real
instruments—to make them believable. To do this you must be
familiar with the sound of real instruments.
It’s like an artist trying to draw a still-life as realistically as possible.
The artist compares the drawing to the real object, notes the difference, and then modifies the drawing to reduce the difference.
When you’re striving for a natural sound, compare the recorded
instrument with your memory of the real thing. How does it sound
different? Turn the appropriate knob on the console that reduces
the difference. You’re creating an illusion of accuracy. To make a
recorded instrument sound real—like it’s really playing in front of
you—you often need to exaggerate things and add some processing.
Alternatively, when you’re mixing, imagine that you’re creating a
sonic experience between the monitor speakers, rather than just
reproducing instruments. Sometimes you don’t want a recording
to sound too realistic. If a recording is very accurate, it sounds like
musical instruments, rather than just music itself.
This approach contradicts the basic edict of high fidelity—to
reproduce the original performance as it sounded in the original
Mixer Operation CHAPTER 12
environment. Some songs seem to require unreal sounds. That way,
you don’t connect the sounds you hear with physical instruments,
but with the music behind the instruments—the composer’s dream
or vision.
Here’s one way to reproduce pure music rather than reproducing
instruments playing in a room: Mike closely or record direct to avoid
picking up studio ambience. Then add reverb. Also add EQ, doubletracking, and effects to make the instrument or voice slightly unreal.
The idea is to make a production, rather than a documentation—a
record, rather than a recording.
Try to convey the musician’s intentions through the recorded sound
quality. If the musician has a loving, soft message, translate that into
a warm, smooth tone quality. Add a little mid-bass or slightly reduce
the highs. If the musical composition suggests grandeur or space,
add reverberation with a long decay time. Ask the musicians what
they are trying to express through the music, and try to express that
through the sound production as well. Listen to the lyrics to see if
the song is intimate (suggesting a warm, dry sound) or a proclamation (maybe with lots of reverb).
Try to keep the mix clean and clear. A clean mix is uncluttered; not
too many parts play at once. It helps to arrange the music so that
similar parts don’t overlap. Usually, the fewer the instruments, the
clearer the sound. Mix selectively, so that not too many instruments
are heard at the same time. Have guitar licks fill in the holes between
vocal phrases, rather than playing on top of the vocals.
In a clear-sounding recording, instruments don’t “crowd” or mask
each other’s sound. They are separate and distinct. Clarity arises
when instruments occupy different areas of the frequency spectrum.
For example, the bass provides lows; keyboards might emphasize
mid-bass; lead guitar may provide upper mids, and cymbals fill in
the highs.
Often the rhythm guitar occupies the same frequency range as the
piano, so they tend to mask each other’s sound. You can aid clarity
by equalizing them differently. Boost the guitar at, say, 3 kHz, and
boost the piano around 10 kHz. Or pan them to opposite sides.
More on judging sound quality is in Chapter 14.
275
276
CHAPTER 12 Mixer Operation
Changes During the Mix
It’s rare to do a mix in which you set the faders and leave them there.
Often you need to change fader levels, EQ, or effects send levels during a mix.
Level changes during the mix should be subtle, or else instruments
will “jump out” for a solo and “fall back in” afterwards. Set faders to
preset positions during pauses in the music. Nothing sounds more
amateurish than a solo that starts too quietly then comes up as it
plays. You can hear the engineer working the fader. If you need to
reduce the level of a loud passage, do so at the end of the preceding
soft passage before the loud one begins.
If you need to change fader levels during the mix, and you don’t
have automation, you might mark these levels next to each fader on
a thin piece of tape. Make a cue sheet that notes the mixer changes at
various counter times. For example:
0:15 Unmute vocal
1:10 Lead solo ⫺5
1:49 Lead ⫺10
2:42 Synth EQ ⫹6 at 12 K
3:05 Start fade, out by 3:15.
Most recording software lets you automate the mix. Details are later
in this chapter under the heading Recording Software with Automated
Mixing.
Record or Export the Mix
When you’re happy with the mix and recording levels, record the mix
on your hard drive with recording software. If you’re mixing with a
DAW, export or save the mix as a 24-bit WAVE file.
If you don’t have automation, and the mix is very difficult, you can
record it a section at a time and then edit the sections together.
To fade out the end of the tune with a hardware mixer, pull down
the mixer master faders slowly and smoothly. In a DAW, either
(1) select all the tracks and insert a fade-out at the end of all the
tracks or (2) export the mix to a stereo file, import the stereo mix
file, and insert a fade-out at the end of the mix.
Mixer Operation CHAPTER 12
Use a fade-out that starts quickly and ends slowly. Try to have the
music faded out by the end of a musical phrase. The slower the
song, the slower the fade should be. The musical meaning of a fade
is something like, “This song is continuing to groove, but the band
is leaving on a slow train.” You might want to postpone fades until
mastering several songs to make an album.
The mixdown is complete. You might record several different mixes
of one song, then choose the best mix. It’s common to record a
mix with the lead vocal up 1 dB, and another with the lead vocal
down 1 dB.
Repeat these mixdown procedures for the rest of the good takes. Give
your ears a rest every few hours! Otherwise, your hearing loses highs
and you can’t make correct judgments.
After a few days, listen to the mix on a variety of systems—car speakers, a boom box, a home system. The time lapse between mixdown
and listening will allow you to hear with fresh ears. Do you want to
change anything? If so, make it right. You’ll end up with a mix to be
proud of.
If you lack good multitrack recordings with which to practice mixing, go to www.raw-tracks.com. There you can download individual
tracks in WAVE or MP3 format, or purchase a CD of raw tracks.
SUMMARY
The following are summaries of the procedures for recording, overdubbing, and mixdown. Use these steps for easy reference.
Recording
1. Turn up the headphone or monitor volume control. Monitor
the aux bus that you’re using for the monitor mix. If your
multitrack is wired to the insert jacks, monitor the stereo bus
instead. If you’re using a DAW, monitoring is automatic.
2. Assign instruments to tracks. To record one instrument per
track, connect its direct-out (or insert send) to a track input.
If you’re using a DAW, select each track and assign it an input
signal from your audio interface.
277
278
CHAPTER 12 Mixer Operation
3. Turn up the input faders, submaster, and master faders to
design center (the shaded portion of fader travel, about threequarters of the way up).
4. Adjust the input attenuators (trim) to set submixes and recording levels. If you’re recording with a DAW, use the volume
controls of the audio interface (they might be in a software
application).
5. Set the monitor mix.
6. Record onto the multitrack recorder.
Overdubbing
1. Assign the instruments or vocals to be recorded to open tracks.
An open track is blank or has already been bounced.
2. Turn up the monitor/cue system.
3. Turn up the submasters and master to design center.
4. Play the multitrack recording and set up a cue mix of the
already-recorded tracks.
5. While a musician is playing, adjust the input attenuation and
recording level.
6. Set the monitor/cue mix to include the sound of the instrument or vocal being added.
7. Record the new parts on open tracks.
8. Punch in and comp tracks as needed.
Mixdown
1. If necessary, set the input selectors to accept the multitrack
recorder output signals.
2. Monitor busses 1 and 2 (or the stereo mix bus). Monitoring is
automatic with a DAW.
3. Assign tracks to busses 1 and 2 (or the stereo mix bus, or the
stereo output of your audio interface).
4. Turn up the master fader to design center. In some mixers, the
submasters should also be up.
5. Set preliminary panning.
6. Set a rough mix with the input faders. Maybe start with all faders at ⫺12 dB, then adjust from there.
7. Set EQ and effects.
Mixer Operation CHAPTER 12
8. Perfect the mix and set recording levels. Set up automation if
your system has it (described next).
9. Record the mix onto a hard drive using recording software. If
you have a DAW, export or save the mix as a WAVE file.
AUTOMATED MIXING
A multitrack mixdown is often a complicated procedure. It can be
difficult to change the mixer settings correctly at all the right times.
So you might want to use automated mixing—have the recorded session file store and reset the changes for you.
As you mix a song, you might adjust the mixer controls several
times. For example, raise the piano’s volume during a solo, then
drop it back down. Mute a track to reduce noise during pauses in
the performance. An automated mixing system can remember your
mix moves, and later recall and reset them accordingly each time
you play back the mix. You can even overdub mix moves; for example, do the vocal-fader moves on the first pass, drum moves on the
second pass, and so on. You also can punch-in fader moves to correct them. Effects changes can be automated as well in some units.
Automated mixing is a feature in digital mixers, software DAWs, and
many recorder–mixers.
Automated mixing has many advantages. With it you can:
■
■
■
■
Perform complicated mixes without errors
Fine-tune the mix moves
Recall mixes weeks or months after storing them, without having to reset the mixer manually each time
Listen to the mix without the distraction of having to adjust
faders
Types of Automation Systems
Three types of automation systems are
1. Automated mixer with non-motorized faders
2. Automated mixer with motorized faders (Flying Faders)
3. Recording software with automated mixing
Each of these is worth a closer look.
279
280
CHAPTER 12 Mixer Operation
AUTOMATED MIXER WITH NON-MOTORIZED FADERS
Suppose you have mixer with non-motorized faders. The mix is
playing and you are adjusting the faders. As you set the position of
each fader, this action produces a MIDI signal that is recorded by a
sequencer. This sequencer is either built into the mixer or is external.
When the mix plays back, so does the sequence of mix moves.
MIDI signals from the sequencer set the volume level for each channel. This is done by varying the gain of a voltage controlled amplifier (VCA) or digitally controlled amplifier (DCA) in each channel.
Because the faders do not move as the mix plays back, the fader positions do not represent the mix you are hearing.
AUTOMATED MIXER WITH MOTORIZED FADERS
This system works as follows. First, you adjust the faders to set up a
mix. This generates MIDI signals that are recorded by a sequencer.
When you play back the mix, the sequencer makes the motorized
faders move up and down as if controlled by a ghost, matching your
mix moves. The position of the mixer faders show the mix levels.
VCAs aren’t needed, which is a benefit because VCAs can degrade the
audio slightly.
RECORDING SOFTWARE WITH AUTOMATED MIXING
Suppose you use a computer DAW to record tracks and edit them.
The monitor screen shows virtual faders that you adjust with a
mouse or control surface. As you set up a mix, the automation feature in the DAW remembers your fader moves, EQ, panning and
plug-in settings. All those moves and settings are reproduced each
time you open a project and play it back.
Automation is easy in DAWs that let you insert a volume envelope
on a track. This envelope is a graph of a track’s fader setting versus
time (Figure 12.3). You might prefer to automate the mix by tweaking a track’s volume envelope on screen instead of moving the fader.
As the song plays, the fader for that track will move up and down,
following the envelope. Many DAWs let you automate other parameters as well, such as reverb-send level, panning, and so on.
What if a track needs a major change in EQ or reverb at a certain
point? Find the section in the track that needs different EQ or reverb,
Mixer Operation CHAPTER 12
FIGURE 12.3
In a track, a volume envelope controls the track’s volume in the mix over time.
create a clip of that section, and move it to another track with those
plug-in settings. Or simply insert a new plug-in into the clip (if your
DAW has that feature.)
As we said earlier, a MIDI/audio recording program includes both
digital audio tracks and sequencer tracks. You automate the audio
track levels by adjusting their gain (fader levels or volume envelopes). You automate the MIDI tracks by adjusting their MIDI volume or key-velocity scaling.
Snapshot versus Continuous Automation
Two types of mixer automation are snapshot and continuous
(dynamic). With snapshot automation, you push a button to take a
“snapshot” or “scene” of the mixer settings. The sequencer stores the
snapshot as MIDI “program changes.” To reset the mixer to any of
281
282
CHAPTER 12 Mixer Operation
these stored settings, you punch-up the appropriate number (MIDI
program change) as the song plays. Recording software also can
record and play snapshots of the mixer settings.
With continuous or dynamic automation, the sequencer records the
motion of the mixer controls—not just their position. Continuous
automation costs more than snapshot and consumes more memory, but permits finer resolution of mix moves. Some digital mixers
do both snapshot and continuous automation with their internal
sequencer. Other mixers require an external MIDI sequencer to
record continuous automation moves.
Some automated mixers let you specify a fade time between snapshots. When the sequencer changes from one snapshot to the next,
the sequencer fades or adjusts the control settings gradually between
the two snapshots. This acts like continuous or dynamic automation.
Automated Mixing Procedure
Here is a suggested order of steps in doing an automated mix:
1. Set up a rough mix manually. You might prefer to start with a
song section where all the instruments are playing.
2. Record the mutes. Mute tracks that aren’t being used, or that
don’t add to the arrangement. For example, a chorus might
sound better without the horns. Unmute each muted track just
before you want it to be heard. If you’re using a DAW, you have
already deleted unwanted portions of each track.
3. Record the fader moves or, in a DAW, draw a volume envelope
for each track. For example, turn up the guitar during the solo,
turn up the toms during the fills, turn down the background
vocals when they get too loud. Save your changes often.
4. Record the effects changes. You might bring up the chorusing
on the acoustic guitar during the bridge, and so on.
5. Sit back and listen to the mix. Is there anything you want to
change?
Another way to do an automated mix is with snapshots. At several
points in the song where you need mutes or fader changes, set the
mute or fader level on the appropriate track and take a snapshot.
Mixer Operation CHAPTER 12
LO-FI RECORDING: HOW TO TRASH YOUR TRACKS
It’s an exciting time for home recording. We have low-cost digital
recorders and mixers, soft synths, and good-sounding cheap mics.
All these tools make it affordable to record with sonic purity and
accuracy. That’s a wonderful development.
But now that anyone can record high-quality sounds, it isn’t such a
big deal anymore. Sure, there’s a place for clean, accurate recordings.
But in many of today’s records, you’ll hear instances of lo-fi sound:
fuzzy vocals, tinny drums, and humming guitar amps.
Lo-fi is the opposite of hi-fi. High fidelity recordings sound like
real musical instruments. Technically, hi-fi sound implies a flat frequency response, no noise or distortion, and low wow and flutter.
In contrast, lo-fi sounds might have a narrow frequency response
(a thin, cheap sound), and might include noise such as hiss or
record scratches. They could be distorted or wobbly in pitch.
Also, the lo-fi attitude snubs its nose at “clean” recordings. Clean
can mean a lot of things: free of noise, free of leakage, free of excess
room reverb. So a lo-fi sound might have tape hiss, some leakage,
and some room acoustics as part of the mix.
Lo-fi really took off with rap music, in which the drum sound was
the opposite of the usual polished studio sound. Instead of a tight
kick, we heard a boomy kick. Instead of a wide-range snare with a
full thump and crisp attack, rap gave us a tinny, trashy snare that was
all midrange.
Want to put some lo-fi sonic interest in your mixes? Let’s look at a
few ways to trash those clean sounds that are so easy to record.
Lo-Fi Frequency Response
A flat frequency response in a mic or mixer tends to give an accurate,
natural sound. If the response is flat, all the musical fundamentals
and harmonics are reproduced in their original proportions. So the
recording sounds accurate or hi-fi.
You can easily make a lo-fi effect by messing up the response. Cut
the highs and lows, boost the mids. Or create a raggedy response
283
284
CHAPTER 12 Mixer Operation
with lots of bumps and dips. Some ways to do this are with EQ, mic
choice, and mic placement.
Play a snare-drum track through your mixer, and turn down the LF
EQ and HF EQ. Boost around 1 kHz or nearby frequencies. Your
snare sound will change from high-budget to bargain-basement.
Beck’s “Soul-Suckin’ Jerk” from the EP Loser is a good example of a
lo-fi drum set.
Or try recording a child’s drum set with its small heavy cymbals and
boomy kick drum. You might loop a hi-hat beat made from this set,
and mix it with a full-range recording of a quality drum set.
Find a toilet-paper tube, or a flexible plastic tube that extends gutter downspouts. Put the tube in front of a mic and sing through the
tube. The resonances in the tube will color the sound in a wild way.
You might track down some cheap old mics at a garage sale, on eBay,
or from vintage mic collectors. Record a few tracks using those mics.
Their frequency response tends to be a complex series of peaks and
valleys that you can’t duplicate with EQ.
Unusual mic placements are fun. Record a guitar amp or vocal with
the mic placed in a wastebasket (Figure 12.4). Hit a cymbal with a
cheap mic while recording its signal. Mike a snare drum from underneath for a thin, zippy effect. If you pick up a crash cymbal off its
edge, the sound will waver as the cymbal tilts when it is struck.
Distortion
Distortion is the addition of harmonics that didn’t exist in the original sound. Here’s an obvious way to create distortion: drive a piece
of recording gear at very high levels—beyond what it can handle. For
example, record drums on a cassette recorder with the meters pinning. Or yell into a “bullet” type harmonica mic so that the mic
distorts. In a DAW, use a distortion plug-in such as iZotope Trash
shown in Figure 12.5 (www.izotope.com).
Another method to add distortion to a track: feed the signal through
an effects box with a distortion setting. Run a drum track through a
guitar stomp box, or through a broken vintage compressor. Feed a
vocal through an Amp Farm or Line 6 Pod processor (Figure 12.6).
You might record some instruments on a cheap cassette recorder
Mixer Operation CHAPTER 12
FIGURE 12.4
The old mic-in-a-wastebasket trick.
FIGURE 12.5
iZotope Trash.
285
286
CHAPTER 12 Mixer Operation
FIGURE 12.6
Line 6 Pod.
(Figure 12.7). The Rolling Stones did that to create the beginning of
“Jumping Jack Flash.”
Noise
Try iZotope’s Vinyl simulation plug-in to add record scratches and
other noises. Another way to have noises in your mix is to record
noisy instruments! You might find an old, creaky upright piano and
do the keyboard tracks with it. Or use an old guitar amp with blown
tubes to record a sound full of hum and hiss.
Leakage
If your mixes are too sterile or studio-clean, you might want to
record some leakage. Leakage (or bleed or spill) is the pickup of an
instrument by another instrument’s microphone. For example, if a
guitar mic picks up the drums from across the room, that pickup of
the drums is called leakage. Since the guitar mic picks up the drums
at a distance, the leakage changes the recorded sound of the drums
from tight to muddy.
Mixer Operation CHAPTER 12
FIGURE 12.7
Recording with a cheap cassette recorder.
Normally we try to isolate instruments or mike close to prevent leakage. But if you want to add some creative leakage to your mixes,
mike a little farther away than you normally do, and record all the
instruments at once.
Room Sound
In the quest for clean, tight recordings, it’s standard practice to partly
cover the walls of the studio with sound-dampening material. This
reduces early reflections, which are less than about 20 msec after the
direct sound from the instrument being recorded. Those early reflections tell the ear that the instrument was recorded in a small room.
Normally we get rid of those early reflections and replace them with
artificial reverb. But a lo-fi recording often includes the sound of the
room as part of the sound of the recorded instrument.
287
288
CHAPTER 12 Mixer Operation
To pick up room reflections, mike farther away than usual and leave
the walls uncovered. Use the room for its coloration, rather than
rejecting the room.
Want to make a synth track more organic and spacious? Run the
synth track through a guitar amp, and mike the amp in stereo from
several feet away. The guitar amp rolls off the highs and adds distortion as well. You could even record a vocal or harmonica through
the amp this way.
For a spacious effect, consider recording several instruments in stereo
with two mics. Pick up instruments or vocals in a hallway, a bathroom, a box, or even outdoors.
Lo-Fi Aesthetics
It’s common to include hi-fi sounds along with lo-fi sounds in the
same mix. This makes a statement to your listeners: “I can record hifi sounds, but I’m choosing not to. The trashy sounds are due to a
conscious choice rather than a lack of recording chops.” If you have
nothing but lo-fi sounds in your mixes, it might sound like you don’t
know what you’re doing—you have no command over the recording
process.
So, your mix might be mostly pristine, full and sparkling, but tainted
with a vocal that sounds like it came through a telephone. The contrast of clean and dirty sounds, modern and vintage, can add a lot of
sonic interest.
The ear delights in complexity. We can add some of that to our
recordings with intriguing, colorful lo-fi sounds.
I recommend the album Mule Variations by Tom Waits. It is a brilliantly creative lo-fi masterpiece. So is Beck’s Timebomb. Others are
“digital hardcore” genre albums by Ronin, Technology Scum, and
Cheap Czad.
CHAPTER 13
Computer Recording
With recording software and a sound card, you can turn your computer into a powerful digital recording studio. This Digital Audio
Workstation (DAW) lets you record dozens of audio tracks, edit them,
add effects, do a mixdown with automation, and burn a professionalquality CD—all in your computer. The cost is only a few hundred
dollars.
In addition to recording audio, most DAW software can act as a
sequencer by recording MIDI data. You can record, edit, and play
both audio and MIDI tracks in the same program.
Another function of a DAW is to edit tracks that were originally
recorded on a hardware multitrack recorder. You can transfer eight
or more tracks at once to your computer by using a sound card
with Alesis Lightpipe or Tascam TDIF connectors, or with an Alesis
FirePort that works with an Alesis HD24 hard-disk recorder.
A DAW has three parts (Figure 13.1):
1. A fast computer with lots of memory and a large hard drive.
2. An audio interface to get audio and MIDI into and out of your
computer.
3. Recording software.
You also need some powered monitor speakers and at least one
mic. If you’re using a sound card that has a low-quality mic preamp,
you’ll also need a good mic preamp or small mixer. Some optional
extras are a control surface and DSP cards. These are explained later
in this chapter.
289
290
CHAPTER 13 Computer Recording
MONITOR
HARD
DRIVE
PCI SLOT, USB PORT,
OR FIREWIRE PORT
COMPUTER
KEYBOARD
ALESIS
FIREPORT
AUDIO INTERFACE
SOUND CARD
MOUSE
CONTROLLER OR MIXER
WITH I/O
AUDIO INTERFACES (CHOOSE ONE)
FIGURE 13.1
A computer DAW.
BASIC OPERATION
When you launch the recording software, you see simulated tracks
and recorder transport buttons such as RECORD, STOP, and PLAY.
You also see a mixer with virtual controls: simulated faders, knobs,
buttons, and meters.
Most DAW software has a few windows or views (Figure 13.2). You
view and manipulate the tracks in the Track window, do edits in the
Track or Edit window, and adjust mixer controls in the Mixer window. Each window can be opened to fill the entire screen. Often you
can do all the work in the Track window.
Recording and Playback
Here’s how the system operates. Audio from your mic, instrument,
mixer, or mic preamp goes to the inputs on the audio interface, which
converts the audio to computer data and sends it to the computer.
Computer Recording CHAPTER 13
FIGURE 13.2
An example of DAW software windows.
The software lets you record this data on the computer’s hard drive.
During playback, the recorded data streams from the hard drive into
the interface, which converts the data back into audio at the interface
outputs. Monitor speakers connected to those outputs play the audio.
Audio data on the hard drive is read by an electromagnetic head.
Because the head can be controlled to jump to any location on disk,
it has random access: it can instantly locate any part of the audio
program. As the head is jumping around, the data pieces it picks up
are read into a buffer memory, then read out at a constant rate.
The waveform of the recorded audio appears on your monitor screen
(Figure 13.3). You can zoom out to see the entire program, or zoom
in to see individual samples.
Like a multitrack tape recorder or hard-disk recorder, a DAW has several tracks to record on. You might record a MIDI drum pattern on
track 1, a bass audio signal on track 2, guitar on 3, keys on 4 and 5,
291
292
CHAPTER 13 Computer Recording
FIGURE 13.3
A waveform editing screen (SONAR Producer 7.0.2).
vocals on 6, and so on. During playback those tracks mix or combine into a stereo signal that plays through your audio interface.
Editing
Editing is a wonderful feature of all computer recording programs.
When you edit audio, you select a part of the audio waveform on a
track (or tracks). Then you cut, copy, or paste that audio, or slide it
in time. It’s like a word processor for audio.
Editing lets you clean up noises, replace bad notes with good ones,
fix bad timing, or create a stuttering effect. With editing you can
rearrange song sections or make choruses repeat without having to
record each one. You might fade out the end of a song or crossfade
between two songs. We cover editing in detail later in this chapter
under the heading Editing Tips.
Mixdown
Once all your tracks are recorded and edited, it’s time for mixdown.
Here is the general procedure:
1. Adjust each on-screen fader with your mouse to set the level of
each track until you create a good balance among tracks.
Computer Recording CHAPTER 13
2. Make selections with the mouse to add panning, EQ, compression, and effects to various tracks.
3. Set up automation so that the computer remembers your mix
moves and reproduces them each time you play the mix.
4. Once your mix is perfected, export it to a stereo WAVE file or
AIFF file on your hard drive.
5. Repeat steps 1 to 4 for all the songs in a demo or album.
6. Burn a CD of the finished mixes or convert them to MP3 files.
7. Alternatively, import the mixes one after the other into a stereo
track. Put the mixes in order with some silence between them.
Burn a CD of the ordered mixes.
With this basic understanding, let’s take a closer look at DAW components: the computer, audio interface, and recording software.
THE COMPUTER
Either Mac or PC will give great results with audio software. Just be
sure that the software you want to use is compatible with your computer platform.
To play a lot of tracks and effects in real time, you need a computer
with a fast central processing unit (CPU); lots of RAM; and a large,
fast hard drive. A minimum system would be a 2 GHz CPU, 512 MB
RAM, and an 80 GB hard drive. Two hard drives are faster: one for
system files and programs, and another for audio data. The drive
should be capable of high-sustained transfer rate (thruput). Current
ATA-100 or ATA-133 drives can sustain up to about 40 MB per second, which is fast enough for multitrack recording.
Multitrack audio consumes a lot of disk storage space. Table 13.1
shows the amount of hard-drive space needed for a 1-hour recording
with various recording formats.
If you plan to record a 1-hour album with about four takes per
song, multiply the Storage Needed by four. In addition to the disk
storage for each song’s tracks, you will need about 30 to 200 MB
per song mix, and up to 700 MB for an 80-minute CD album of the
song mixes.
293
294
CHAPTER 13 Computer Recording
Table 13.1
Hard-Drive Storage Required for a 1-Hour Recording
No. of tracks
Bit depth
Sampling rate
Storage needed
2
16
44.1 kHz
606 MB
2
24
44.1 kHz
909 MB
2
24
96 kHz
1.9 GB
8
16
44.1 kHz
2.4 GB
8
24
44.1 kHz
3.6 GB
8
24
96 kHz
7.7 GB
16
16
44.1 kHz
4.8 GB
16
24
44.1 kHz
7.1 GB
16
24
96 kHz
15.4 GB
24
16
44.1 kHz
7.1 GB
24
24
44.1 kHz
10.7 GB
24
24
96 kHz
23.3 GB
AUDIO INTERFACE
Once you have a fast computer with a large hard drive, you need
a way to get audio signals into and out of the computer. An audio
interface does the job. Four types of interface are listed below, and
we’ll look at each one (see Figure 13.1).
1. Sound card (also called a PCI audio interface)
2. FireWire or USB audio interface
3. Control surface with I/O (Input/Output connectors)
4. Alesis FirePort. This converts WAVE files from an Alesis HD24
FST-formatted hard drive to FireWire, and sends the data to
your computer. FireWire (IEEE-1394) is a high-speed data link
between digital devices.
Sound Card (PCI Audio Interface)
The simplest form of interface is a 2-channel sound card, which
plugs into a PCI user slot in your computer’s motherboard. Low-cost
sound cards have unbalanced 1/8-inch (mini) phone jack connectors, which include a mic input, stereo line input, and stereo line
output. Generally, the sound quality and connectors of low-cost
cards aren’t up to professional standards.
Computer Recording CHAPTER 13
FIGURE 13.4
A sound card.
The next step up is a sound card with 1/4-inch TRS connectors
(Figure 13.4) or XLR connectors on cables. It offers 2 to 8 balanced
inputs. Current high-quality sound cards can record with 24-bit resolution. Many sound cards have MIDI connectors and an onboard
synthesizer. Some have a FireWire port. Examples are sound cards by
RME, Frontier Design, M-Audio, Lynx Studio Technology, E-MU, and
Echo Digital Audio. A comparison of some sound cards is at www
.pcavtech.com/soundcards/.
Choose either a 2-channel or multichannel card. A 2-channel sound
card is adequate if you’re recording one instrument at a time (such
as a vocal, sax, keyboard, guitar, or a mix of several drum mics). You
can overdub more parts using the same stereo input on different
tracks. Also, a 2-channel sound card is sufficient if you’re using an
Alesis FirePort, which copies several Alesis HD24 recorder tracks to
your hard drive, bypassing the sound card. In that case, you’d use the
sound card just for monitoring.
You’ll need a multichannel sound card (or a multitrack hard-disk
recorder) if you want to record several instruments at once, such as a
band or individual mics on a drum set. This type of card has several
connectors on short cables, one per channel. Each instrument’s signal goes to a separate channel in the sound card, and each channel’s
295
296
CHAPTER 13 Computer Recording
signal is recorded on a separate track. You can install two or more
sound cards side by side to get more channels.
Check the sound card’s Web site to make sure that it’s compatible
with your computer and recording software.
Here are some desirable features to look for when shopping for a
sound card:
■
■
■
■
■
24-bit, 44.1-kHz minimum, full duplex recording. A full duplex
card can record and play back simultaneously. The card works
on two DMA channels.
85 dB or greater S/N ratio, 20 Hz to 20 kHz frequency response
(⫾0.5 dB or less).
XLR connectors, 1/4-inch TRS phone jacks, RCA phono jacks,
MIDI connectors.
Operates at ⫹4 dBu as well as ⫺10 dBV.
Comes with drivers that work with your operating system and
your recording software.
If the sound card includes an onboard synthesizer chip, look for
these features:
■
■
■
■
■
■
General MIDI (GM) compatible
MPU-401 MIDI interface compatible
Wavetable synthesis (better sound than FM synthesis)
Programmable synth patches
24-note polyphony minimum
2 MB of wavetable ROM or RAM minimum
Some PCI cards have an I/O box, in which all the connectors are in
a common chassis (see Figure 13.5). Because the interface’s analog
FIGURE 13.5
E-MU 1616M, an example of a PCI audio interface I./O box.
Computer Recording CHAPTER 13
circuits are outside of the computer, they tend to pick up less computer electrical noise than analog sound cards. This interface accepts
analog audio or MIDI signals, converts them into computer data,
and sends this data to the computer via a card in a PCI slot.
FireWire or USB Audio Interface
This is a chassis with 2 to 16 mic preamps, A/D converters, and a
USB or FireWire port that sends digital audio on a single cable to
your computer (Figure 13.6). Some units have MIDI ins and out,
and a high-impedance instrument input for an electric guitar, electric
bass, acoustic guitar pickup, or synth.
A USB or FireWire interface connects to a mating connector on the
outside of your computer. To install a PCI interface you must open
the computer.
Some mixers could be called audio interfaces if they include a
FireWire or USB port. It sends the output signal of each mic preamp
to a separate track in your recording software, and returns the stereo
mix back to the mixer for monitoring. This can be a good system for
recording a band because it offers easy setup of monitor mixes with
effects. Chapter 4 lists some models of audio interfaces and mixers
under the heading Computer Digital Audio Workstation.
Figure 13.7 shows a USB port and cable connector.
FireWire comes in two speeds: FireWire 400 (IEEE 1394) which runs
at 400 Mbps (megabits per second), and FireWire 800 (IEEE 1394b)
which runs at 800 Mbps (100 MB per second). In contrast, USB 2.0
high-speed is 480 Mbps.
FIGURE 13.6
PreSonus FireStudio Project, an example of a FireWire audio interface.
297
298
CHAPTER 13 Computer Recording
FIGURE 13.7
USB port and cable connector.
FIGURE 13.8
FireWire port and cable connector.
You can add a FireWire port to your computer with a $35 FireWire
card, and some computers even have FireWire built in. To prevent
data errors, get a FireWire card with a Texas Instruments (TI) chipset.
Figure 13.8 shows a FireWire port and cable connector.
USB and FireWire devices are hot-swappable: you can insert or
remove the connector while the computer is on. Both formats are
compatible with Mac or PC.
Both ports are also available in PCMCIA cards and CardBus cards,
which fit into laptops. A PCMCIA card is a credit-card-size memory
card or I/O device that connects to a slot in a computer. CardBus is
an advanced PCMCIA card with faster speed due to its direct memory access (DMA) and 32-bit data transfer.
Audio Interface Features
When deciding which interface to purchase, consider the features
below.
DIGITAL I/O
Some interfaces have digital inputs and outputs as well as analog.
Digital I/O comes in various formats (AES3, S/PDIF, Lightpipe),
Computer Recording CHAPTER 13
so make sure that the interface’s format matches that of your digital recorder, digital mixer, or outboard A/D converter. See Chapter 9
under the heading Digital Audio Signal Formats.
ANALOG I/O
Check whether the analog I/O on the interface is balanced or unbalanced. Balanced connections reduce electrical interference picked up
by cables. However, unbalanced connections cost less, and usually
are adequate for cable runs under 10 feet. Balanced connectors are
XLR or 1/4-inch TRS phone jacks. Unbalanced connectors are RCA
phono jacks or 1/4-inch TS phone jacks.
SAMPLING RATE AND BIT DEPTH
Available sampling rates in audio interfaces are 44.1 to 192 kHz.
Recordings made with higher sampling rates may give better sound
quality but consume more hard-drive space. If you’re making a CD
or MP3 file, a sampling rate of 44.1 kHz does the job.
Most interfaces can handle 24-bit as well as 16-bit signals. Recordings
made with 24-bit resolution sound a little smoother and cleaner
(less distorted) than 16-bit recordings. Even if your final product is
a 16-bit CD or MP3 file, it will sound better with a 24-bit recording that is dithered down to 16 bits during mastering. See Chapter 9
under the heading Dither.
MIDI PORTS
Many interfaces have MIDI ports. A MIDI IN port accepts MIDI signals from a MIDI controller and sends them to your computer’s
sequencing software. A MIDI OUT port sends MIDI signals from
your computer’s sequencing software to a synthesizer or sound generator. If your audio interface has MIDI ports, you don’t need a separate MIDI card in your computer.
WORD CLOCK
Some interfaces offer word clock connectors, which send and receive
timing signals for digital audio. Word clock is used to synchronize
the digital devices in a large studio for real-time transfer of digital
audio. See Chapter 9 under the headings The Clock and Jitter.
299
300
CHAPTER 13 Computer Recording
DRIVER SUPPORT
An audio driver is a program that allows recording software to transfer audio to and from an audio interface. Most interfaces are sold
with several types of drivers. Be sure that your interface comes with
drivers that work with your operating system and your recording
software.
A good driver has a minimum latency spec under about 4 msec.
Latency is the signal delay through the driver and interface to the
monitor output. This can be a problem in recording a soft synth,
because you hear the playback of the synth a few msec after striking the keys on the MIDI controller. See Appendix B, Minimizing
Latency.
The most popular audio driver formats are ASIO, WDM, DAE,
Direct I/O, GSIF, MAS, SoundManager, WAVE, and Core Audio.
Here’s a brief description of each format:
■
■
■
■
■
■
ASIO (Audio Streaming Input and Output; Mac OS 9⫹, PC):
A very popular driver developed by Steinberg. Allows multiple
channels of simultaneous input and output, and low latency
with software synthesizers.
WDM (PC): Windows Driver Model with Kernel Streaming, a
multichannel driver. Allows low latency with WDM-compatible
audio hardware and DXi software instruments. DXi stands for
DirectX Instruments, Cakewalk’s virtual instrument integration
standard. DirectX audio effects can be used live on input signals, not just during playback. This lets you monitor and record
effects in real-time as you’re recording.
DAE (Mac, PC): Used only with Digidesign audio interfaces.
It’s a multichannel driver that runs with a compatible host such
as Pro Tools, Logic, and Digital Performer. DAE lets you use
RTAS and/or TDM plug-ins (explained later under the heading
Digidesign Pro Tools).
Direct I/O (Mac, PC): Works with Digidesign interfaces as a
multichannel driver only. Doesn’t let you run RTAS or TDM
effects.
GSIF (PC): Permits very low latency when playing samples from
hard disk with TASCAM’s GigaStudio software sampler.
MAS (Mac): Developed by Mark of the Unicorn (MOTU). Offers
resolutions up to 24/96 and multiple simultaneous input and
Computer Recording CHAPTER 13
■
■
■
■
output channels. It’s also a format for plug-ins (software audio
effects).
SoundManager (Mac): MacIntosh’s standard audio driver. It
lets you record and play mono and stereo files up to 16-bit and
44.1 kHz. Has a moderate amount of latency.
WAVE (PC): The PC standard audio driver. WAVE can be used
with a variety of audio interfaces (like Sound Blaster-type sound
cards) to record and play mono or stereo audio. Has a moderate amount of latency.
Core Audio. Apple’s low-latency driver used in Mac OS X.
MME is an older driver that offers lower performance than
newer ones.
ASIO or WDM are the main choices for PC. Cubase seems to work
best with ASIO; Cakewalk Sonar works with WDM or ASIO. Choose
drivers recommended by (or supplied by) the audio interface
manufacturer.
Caution: If you have multiple drivers installed, they may conflict.
Then the computer might crash or the recording software might not
access the audio interface. Delete unused drivers.
OTHER OPTIONS
Listed below are several features that some interfaces offer.
■
■
■
■
■
■
■
Zero-latency monitoring: The input signal to the interface is fed
to its output, so you hear the input signal without any signalprocessing delay (latency).
Pro Tools compatible: The interface works with Pro Tools
recording software and hardware.
Surround sound: Provides 5.1 or 7.1 surround sound monitoring.
Powered by FireWire bus or USB bus: The FireWire or USB connection powers the interface; you don’t need another power
supply.
Battery powering: This makes the interface portable for on-location recording with a laptop.
Supplied recording software: The interface is packaged with
recording software, so you might not need to buy other software.
SMPTE sync: The interface will synchronize with a SMPTE
timecode signal (explained later in this chapter under the heading Audio for Video.).
301
302
CHAPTER 13 Computer Recording
Control Surface
So far we’ve covered the computer and audio interface. Let’s look at
other hardware for a computer studio.
Using a mouse to adjust the controls on-screen is slow and can lead
to repetitive stress syndrome. An alternative is a DAW control surface or controller (Figure 13.9). It’s a chassis with physical controls
for software functions. Resembling a mixer with real faders, knobs,
and transport buttons, it lets you adjust the software’s virtual controls that you see on the monitor screen. The controller attaches to
the computer through a MIDI connector, Ethernet connector, USB,
or FireWire port.
Many control surfaces also act as audio interfaces: they include mic/
line inputs and outputs (I/O) and MIDI connectors.
Some control surfaces are dedicated to one DAW recording program,
such as Pro Tools Mbox or 003 Series. Others are universal and
FIGURE 13.9
M-Audio Project Mix, an example of a control surface.
Computer Recording CHAPTER 13
they work with several different DAW programs. Be sure to check
whether the controller you want to buy will work with your recording software.
What if your controller has 8 faders, but your project has 16 virtual
faders on screen? You can make the controller access groups (banks)
of virtual faders, 8 at a time, by pressing a bank switch. Normally,
your controller faders will affect virtual faders 1 through 8. Press
the bank switch to make the controller faders affect virtual faders
9 through 16. Press it again to access virtual faders 17 through 24,
and so on. Controllers that have 4 faders can access banks of virtual
faders 4 at a time.
Advanced controllers offer these features:
■
■
■
■
■
■
■
■
Motorized faders: The faders in the control surface are motorized, so they move like the virtual faders on screen.
Stand-alone mixer mode: You can use the control surface as a
regular mixer.
Footswitch jack: Accepts a footswitch for punch-in/out. The
footswitch works only if your recording software supports this
function.
Meters.
Monitoring section.
Aux send/return: Allows you to connect an external analog signal processor.
Insert jacks: An insert jack in series with each channel’s signal
allows you to plug in an analog compressor.
Expandability: You can add more controllers, side by side, to
control more virtual faders at once.
Note that a MIDI controller or a MIDI fader box can control a
sequencer’s MIDI tracks, but not audio virtual controls. A control
surface can control both.
The Frontier Design Group TranzPort is a wireless remote control for
a DAW, letting you control your DAW from the studio. It includes
transport controls, metering, a footswitch jack, and more. See the
Web site www.frontierdesign.com for more information.
The Frontier Design Group AlphaTrack (Figure 13.10) and the PreSonus
Faderport are controllers with several buttons and one long-throw
fader. They are easier to use than a mouse and have a small footprint.
303
304
CHAPTER 13 Computer Recording
FIGURE 13.10
Frontier Design AlphaTrack, an example of a single-fader controller.
Alesis FirePort
This is a special type of audio interface that transfers audio tracks
from an Alesis HD24 multitrack recorder to a computer. The HD24
records tracks as WAVE files on an Alesis FST-formatted hard drive
that is plugged into the recorder. After recording tracks, you can
remove the hard drive and plug it into an Alesis FirePort (Figure
13.1), a small device that converts the WAVE files of those tracks to
FireWire format and sends them to your computer hard drive for
editing and mixing.
As we’ve seen, there are a wide range of features and connectors
among different interface models. A good interface is well worth the
price because of its top-quality sound, convenient connections, and
easy control of audio signals.
Computer Recording CHAPTER 13
DSP CARD
Another DAW option is a digital signal processing (DSP) card that
you plug into a user slot in your computer’s motherboard. Why is
it helpful? The more software effects (plug-ins) you use, the fewer
tracks you can mix, because effects put a big load on the CPU.
Normally this isn’t a problem except in mixes with more than 24
tracks and loads of plug-ins in use. You can resolve this issue by
installing a DSP card, which handles the processing for the plug-ins.
The plug-ins access the card rather than the CPU. You can install
multiple cards to expand the system. Examples include: TC Works
Powercore (www.tcelectronic.com), Digidesign TDM-based Pro Tools
cards (www.digidesign.com), Mackie UAD-1 (www.mackie.com),
and Sonic Core (www.soniccore.com) PCI cards.
ANALOG SUMMING AMPLIFIER
This device is a mixer without faders, used to mix tracks or submixes from a DAW through analog circuitry. Two examples are the
Dangerous 2BUS 16 ⫻ 2 mixer and the Manley Labs 16 ⫻ 2. By
using a summing amplifier, you bypass the DAW’s internal mix bus
which might create slight distortion due to rounding errors, numeric
overload, and bit truncation. (Some DAW software claims not to
produce those errors, such as Pro Tools TDM systems.) You connect
the analog output of each DAW track to a summing amplifier input,
and use the DAW’s faders and automation to set mix levels. Some
engineers claim they avoid the distortion of mixing entirely inside
the computer by recording and mixing at lower levels.
RECORDING SOFTWARE
Now let’s turn from hardware to software. Some popular programs
for recording, editing, and mixing are listed in Chapter 4 under the
heading Computer Digital Audio Workstation. They share a lot of
the same features but implement them differently. Also, each product has unique features and its own GUI (graphical user interface;
Figure 13.11).
Check out the products online to see how you might like working
with them. Some Web sites have interactive demos; others have free
trial versions that you can download. Many sites have tutorials.
305
306
CHAPTER 13 Computer Recording
FIGURE 13.11
Cakewalk Sonar Producer track view.
When you first install recording software or hardware, expect some
frustration. Your software may not be compatible with your hardware, or may not perform as expected. You might need to modify
certain settings in your computer system or recording software. Read
the manual and any readme files that came with your product. Also,
check out the products’ Web sites for knowledge bases, FAQs, etc.
Each DAW program has online discussion groups or forums that can
be very helpful. If you’re still having problems, call or e-mail the tech
support for your product.
For each recording program, books are available that provide operating details and power-user tips. To find them, search www.amazon
.com or the references in Appendix D. Type the name of the program
and manufacturer in the search field.
This chapter isn’t meant to duplicate the information in those specialized books or in manufacturers’ Web sites. Instead, it provides an
overview of recording software.
Computer Recording CHAPTER 13
Features
Here are some features to look for in recording programs:
■
■
■
■
■
■
■
■
■
■
■
■
■
Multitrack digital recording from 2 tracks on up: Some programs can record and play back an unlimited number of tracks
depending on the speed of your computer, RAM, and hard
drive.
MIDI sequencer: This lets you import or record several tracks
of MIDI performance data, edit it, and play it. MIDI sequencer
tracks can be mixed with audio tracks. Some programs include
MIDI effects, MIDI plug-ins, and input quantizing.
Freeze function mixes down effects and edits on a track to
another track with embedded effects. This frees up CPU resources.
Freeze also converts MIDI soft-synth tracks to audio tracks, which
also reduces CPU loading. Unfreeze returns to the original tracks
for editing.
Peak and rms metering.
Compatibility with control surfaces.
Surround mixing (in high-end programs).
Customizable GUI: Change the GUI to suit the way you work.
Create configurations to use as templates for similar projects.
Keyboard shortcuts: Instead of constantly using a mouse, tap
certain keys on your computer keyboard. This speeds operation
and gives your mouse hand a rest.
Virtual tracks: Extra takes of a musical performance that are
recorded on the hard drive. During mixdown you can choose
the best takes to use, or the best parts of each take.
Automated mixing: All settings for a song project can be saved
and automated. The computer remembers your mix moves,
and sets the mixer faders and pan pots accordingly during subsequent mixdowns. Some programs let you automate effects
settings as well. For more information, see the section on
Automated Mixing in Chapter 12.
Loop-based composition, construction, and editing tools,
ACID-loop and MIDI Groove Clip support. All these are covered later in this chapter. Loops and samples are included in
high-end programs.
Locate and marker points: Mark several points in a song (intro,
verse, chorus, solos) so that you can go to them instantly.
Routing or virtual patchbay: Assign any input to any track.
307
308
CHAPTER 13 Computer Recording
■
■
■
■
■
■
■
■
■
■
■
■
■
■
■
■
■
■
■
Automatic pitch correction of audio (by Autotune, Melodyne,
or V-Vocal).
Automatic timing correction of notes by Beat Detective (Pro
Tools) or Audiosnap (Cakewalk Sonar Producer) let you lock
the tempo of audio tracks, such as drums, to a tempo map. You
can also make one track follow another track’s tempo, or extract
timing from audio tracks and apply them to MIDI tracks.
Audio-to-MIDI conversion: For example, have a singer’s notes
play the same pitches on a synthesized cello.
Video display: An on-screen window that shows video clips.
Found in high-end software, this feature lets you synchronize
music and sound effects to a video program.
CD recording: Record your mixes on a CD burner.
PQ editing: Set up a list of CD song-start times before burning a
CD of your mixes.
Import and export various file types (such as WAV, MP3, WMA,
OMF), sampling rates and bit depths.
32- or 64-bit floating-point resolution in the DSP.
Support for 64-bit operating systems and dual-core CPUs.
ASIO and WDM compatibility.
Ability to learn the function of each controller encoder or button, mapping the controller to the on-screen controls.
Spectral analysis: A display of level versus frequency of the
audio program as it progresses in time.
Notation application: Converts a MIDI file to musical notes on
a treble and bass clef.
Sync: Synchronization to SMPTE timecode or MIDI timecode
(described later in this chapter under the heading Audio for
Video).
Support for DirectX and VST audio effects. (DirectX is a standard audio driver.)
Support for DXi and VSTi soft synths. (VSTi is an acronym for
Virtual Studio Technology for Instruments, Steinberg’s virtual
instrument integration standard.)
Included soft synths and samplers.
Convolution reverb, vintage compressors, audio restoration,
mastering tools.
Plug-ins with delay compensation: Described in the next section.
Computer Recording CHAPTER 13
Plug-Ins
As we described in Chapter 10, a plug-in is a software module that
adds digital signal processing or effects to a DAW recording program.
Recording software comes with several plug-ins already installed.
You also can purchase and download plug-ins from the Web. Unlike
hardware effects, plug-ins are instantly upgradeable—just download
the latest version.
When you click on the name of a plug-in, a screen pops up that
looks like a hardware processor with knobs, faders, lights, and
meters. Parameters (such as reverb time, chorus depth, or compression ratio) can be adjusted, and most parameters can be saved and
automated.
A number of plug-ins can simulate or model the sound of specific
devices. For example, Amp Farm by Line 6 is a guitar-amp modeling
plug-in, while Antares Microphone Modeler can simulate the sound
of popular studio mics. Vintage compressors, tube mic preamps,
concert halls, and analog tape recorders have been modeled as well.
Some plug-ins run in real-time as the audio is playing. Others process a sound file and create a new file as with normalization or time
compression/expansion.
Plug-in effects can be applied in three ways:
1. Master effect: The effect is on the master output bus, and processes the entire mix. Multiband compression is a typical master effect.
2. Aux-send or loop effect: The effect is on an aux bus. Set the
plug-in’s dry/wet mix control all the way to wet or 100% mix.
Adjust the amount of effect on each track by turning up the
track’s aux send control. This option reduces the number of
plug-ins needed for a mix, and so reduces the load on your
CPU. Delay effects—reverb, echo, chorus, and flanging—are
aux effects.
3. Insert effect: This effect is on a specific track and affects only
that track. EQ and compression are insert effects. If you insert
a delay effect into a track, you adjust the dry/wet mix control
in the effect to control the amount of effect that you hear.
309
310
CHAPTER 13 Computer Recording
Plug-ins come in several formats; the most popular for Windows are
Steinberg’s Virtual Studio Technology (VST) and DirectX. DirectX is
a group of application-program interfaces that enhance multimedia
(video and audio) on Windows systems.
You can find plug-ins on their manufacturers’ Web sites or on the
Web sites of recording-software companies. For example, Digidesign’s
Web site lists dozens of plug-in partners. Another resource is www
.kvraudio.com.
Digidesign Pro Tools
Pro Tools is the industry standard DAW system. However, other
excellent programs are available. People often use the term “Pro
Tools” generically to mean “digital audio workstation.” Because Pro
Tools is so popular, and has a number of configurations that can be
confusing, I’ll describe them here. There are three basic categories:
Pro Tools LE, M-Powered, and HD.
Pro Tools LE and M-Powered are 32-track recording systems that use
your computer’s CPU for all processing. Your CPU speed, hard-drive
speed, and amount of RAM determine how many plug-ins and soft
synths you can use simultaneously.
Pro Tools LE software has nearly the same user interface as the professional Pro Tools HD systems (Figure 13.12). It’s intended for
singer/songwriters, project and personal studios, remote recording
engineers, and bands. Pros can use it at home because it is compatible with the high-end Pro Tools HD system. Cost is $500 to $3000.
LE components include Pro Tools LE software and a choice of audio
interface: Mbox 2 Mini, Mbox, Mbox 2 Pro, 003 Rack Factory, 003
Factory control surface, or Command|8 USB control surface.
■
■
The Mbox series are USB or FireWire audio interfaces with one
or two Focusrite mic preamps and various I/O options.
003 Factory is a control surface and audio interface that communicates with your computer via FireWire. It includes 8 analog ins
and outs, 4 mic preamps, 8 channels of ADAT I/O, stereo S/PDIF
and MIDI I/O, 8 touch-sensitive faders, 8 motion-sensitive rotary
encoders, transport buttons, and automation. Any changes you
make in your Pro Tools software are duplicated on the control
surface and vice versa.
Computer Recording CHAPTER 13
FIGURE 13.12
Pro Tools LE screen view. Source: www.en.wikipedia.org.
■
■
003 Rack Factory is 003 Factory in a 2U rack-mount chassis,
without the control surface.
Command|8 is a USB control surface with 8 bankable motorized faders, rotary encoders, transport controls, and displays.
Pro Tools M-Powered™ software is designed to work with certain
M-Audio® audio interfaces and control surfaces. You can also use
other Pro Tools products such as the Command|8® control surface. M-Powered software includes many of the same features as
the professional LE and HD systems, and M-Powered sessions will
open in those other systems. Easy to use, the software works with
Windows XP and Vista, and Max OS-X computers. The software costs
$299 while the interfaces range from $100 to $500. The M-Audio
ProjectMix I/O control surface sells for $1249.
Pro Tools HD is a high-resolution professional system with DSP
cards, up to 192 kHz sample rate, expandable I/O, and many options.
It comes in three Core systems (each with more DSP power and
more I/O) and one or more audio interfaces. Each system requires
the Pro Tools 192 I/O or the 96 I/O audio interface, and supports
311
312
CHAPTER 13 Computer Recording
legacy Pro Tools audio interfaces for additional inputs and outputs.
Cost is $10,000 to $125,000 or more.
Pro Tools HD Systems offer dedicated DSP cards that free up your
CPU for other tasks. The result is you can run more tracks, more plugins, and more soft synths at the same time. HD offers a wide range
of sample rates, hi-res audio, and up to 192 tracks. An HD system
can have 8 to 96 ins and outs. The three basic Core systems have one,
two, or three DSP cards, and a system can use up to seven DSP cards.
Listed below are the three Core systems. Each can be expanded by
adding more HD Accel cards.
■
■
■
Pro Tools HD 1 includes the HD Core DSP card, which handles
up to 32 channels of I/O and up to 96 tracks.
Pro Tools HD 2 Accel includes the HD Core card and an HD
Accel card, which provide more than four times the mixing and
processing power of HD 1 systems and 64 channels of I/O and
192 tracks.
Pro Tools HD 3 Accel, the most powerful system, includes the
HD Core card and two HD Accel cards, which provide up to 96
channels of I/O and 192 tracks.
Digidesign Pro Tools HD systems also include these interface options.
■
■
■
The 96 I/O Audio Interface converts audio at up to 24-bit/
96 kHz. It includes 8 channels of analog I/O, 8 channels of
ADAT optical I/O, 2 channels of AES/EBU and S/PDIF I/O,
and word clock. It can link to other 96 I/O interfaces and to
Digidesign 888/24, 882/20, 1622, or 24-bit ADAT Bridge.
The 192 I/O Audio Interface converts audio at up to 24-bit/
192 kHz. It includes 8 channels of analog I/O, 8 channels of
AES/EBU, 8 channels of TDIF, 16 channels of ADAT, and 2 additional channels of ADAT or S/PDIF digital I/O. A 192 Digital
expansion card provides 8 more channels of AES/EBU, TDIF,
and ADAT I/O connections and a Legacy port to link to older
Digidesign interfaces.
The ICON D-Control worksurface is a large mixing console/
controller surface designed to work with Pro Tools software.
C|24 is another powerful control surface.
New features of the latest Pro Tools software are on the www.protools
.com Web site.
Computer Recording CHAPTER 13
Plug-ins for Pro Tools come in four formats: TDM, HTDM, RTAS
(Real Time Audio Suite), and AudioSuite. TDM plug-ins work only
with Pro Tools’ professional TDM systems and use dedicated cards
that plug into your computer. HTDM is the same for the Pro Tools
HD (high definition) system. Many RTAS plug-ins are the same
as TDM plug-ins, except that they use your computer’s processor
instead of specialized TDM DSP cards. All the above plug-ins work
in real time.
AudioSuite plug-ins don’t work in real time. Instead, they process
a track’s signal and create a new track in which the effect is part of
the recorded sound; that is, the effects are embedded in the track.
Playing a track with embedded effects is less work for the CPU than
playing a track with real-time effects. Less load on the CPU means
you can play more tracks without drop-outs.
Loads of plug-ins are available for Pro Tools. Many are included, and
some can be downloaded from third-party vendors. Use the plug-in
finder on the Pro Tools Web site (www.protools.com).
Included with every Pro Tools system is a Music Production
Enhancement Suite, which includes soft synths, looping, guitar-amp
modeling, a sample player and tube EQ.
OPTIMIZING YOUR COMPUTER FOR MULTITRACK
RECORDING
Once you’ve chosen some recording software and installed it, you’ll
want to make your computer run as fast and glitch-free as possible.
You need to optimize its settings for best results. Appendix B suggests some ways to speed up your hard drive, reduce software interruptions, and reduce the CPU usage. If you follow those tips, you
will have a faster system that handles more plug-ins and more tracks
at once. Also, when you play tracks or burn CD-Rs, clicks and dropouts in the audio will be less likely.
USING A DAW
Let’s say you purchased some DAW hardware and software, and
tweaked your computer to make it fast and reliable. It’s ready to rock.
Here are some tips on setting up and using your DAW.
313
314
CHAPTER 13 Computer Recording
POWER AMPLIFIER
AND MONITOR SPEAKERS,
OR POWERED MONITORS
COMPUTER
LINE OUT
MIC
IN
AUDIO
INTERFACE
MIC OR DI
OR
MIC OR DI
LINE
IN
SOUND CARD,
AUDIO INTERFACE,
CONTROLLER OR MIXER
MIC
PREAMP
USB,
FIREWIRE
OR
PCI SLOT
HEADPHONES
FIGURE 13.13
Connecting audio equipment to a DAW.
Connections
First, connect your audio equipment to the DAW as shown in Figure
13.13. You may need some adapter cables between devices. The following adapter cable connects a sound card’s unbalanced I/O to a
power amp, powered speakers, or mixer: Stereo 1/8-inch phone plug
to two unbalanced phone plugs, or two RCA phono plugs (whatever
matches your equipment).
Software Settings
Starting from your computer’s desktop view, select Start ⬎ Settings ⬎
Control Panel. Double-click Sound and Audio Devices ⬎ Audio tab.
Set the Playback and the Recording default devices to the sound card
or interface you’re using for audio editing.
Before you record, set a bit depth and sampling rate for the project,
such as 24-bit/44.1 kHz.
If you want to record from a mic or instrument, you’ll record their
amplified signal on an audio track. If you want to record from a
MIDI controller, you will record its MIDI signal on a MIDI track.
Details on recording MIDI sequences are in Chapter 16.
Set up the audio drivers in your recording software. This setting might
be called “Record timing master” and “playback timing master.” Specify
Computer Recording CHAPTER 13
the input and output devices for each audio track, such as “sound-card
left channel input” and “sound-card stereo outputs.” Also, specify the
input and output devices for each MIDI track, such as “sound-card
MIDI IN” and “sound-card stereo outputs.”
If you’re recording many tracks simultaneously, or at a high sample
rate, this may generate more data than your CPU and buffers can
handle—causing drop-outs. A drop-out is a short silence or a noise
burst (glitch) in the recorded audio. To prevent drop-outs while
recording, turn off the track playback metering, effects, and waveform preview. Increase buffer size. Turn off or reduce video acceleration. Don’t zoom or scroll while recording. Also, follow the tips in
Appendix B on optimizing your computer for multitrack recording.
The buffer setting is critical. A large buffer lets you use many tracks
and plug-ins without drop-outs, but it creates a lot of latency (monitoring delay). If latency is high, and you press a key on a MIDI
keyboard, a soft synth will respond after a noticeable delay. A small
buffer minimizes latency so soft synths respond faster, but can cause
drop-outs. Generally it’s best to set the buffer/latency small (under
4 msec) while tracking soft synths and set it large otherwise (maybe
25 msec). See Appendix B under the heading Minimizing Latency.
To monitor a live mic signal without latency while overdubbing, listen to a mix of the live mic signal and the previously recorded tracks.
Mute the mic input signal in the sound card’s software and disable
Echo Input Monitoring in the mic’s track.
Editing Tips
Now that your DAW is set up, you can record and edit audio. When
you select a portion of the audio program, you create a clip or region.
Examples of clips are an entire song, the chorus of a song, a guitar
track, a drum riff, or a single note. Using a mouse, you can create a
clip by marking its beginning and end points in the waveform. In
Figure 13.14, a clip is selected and highlighted.
A clip is actually a pair of pointers to part of an audio file on your
hard drive. One pointer is the data address for the beginning of
the clip, and the other pointer is the address for the end of the clip.
Rather than containing audio data, clips tell the software which
section(s) of the audio file to play.
315
316
CHAPTER 13 Computer Recording
FIGURE 13.14
A clip is selected and highlighted.
When you set up the sequence of clips in a mix, you’re telling the
hard-drive head which pointers to play in what order. Or when you
delete a clip, you’re telling the hard-drive head to skip the clip’s
pointers during playback. If you copy an audio clip, the software
doesn’t make copies of the audio file it points to—instead, it plays
the same audio file each time it sees a clip pointing to that file. Or
when you split a clip into parts and put them in a different order,
the audio file itself isn’t split. Instead, the software plays the sections
of the audio file that the clips point to, in the desired order.
These types of edits are called nondestructive edits. Only the pointers
change; the data on disk are not changed or destroyed. Nondestructive
edits aren’t permanent. If you don’t like an edit, you can undo it and
try it again.
Some types of edits or processing are destructive: they write over the
data on the disk. However, some recording programs save the data
before you edit it. Then you can undo the change by reverting to the
saved data.
I’ll give examples of several things you can do with editing.
Computer Recording CHAPTER 13
FIGURE 13.15
A guitar track with a noise deleted.
DELETE NOISES
Suppose you recorded an acoustic guitar track, but it has a chair
squeak between the verse and chorus. Zoom in on that noise, play it,
highlight it, and delete it (type Ctrl-X with most DAWs). You might
have the option to close up the space you created—don’t do that
in this case. Figure 13.15 shows a guitar track with a noise deleted
between notes.
CLEAN UP TRACKS
Before the beginning of a recorded song, you’ll typically hear noises
and a count-off. To remove them, define the group of noises as a clip
and delete it (Ctrl-X). Some software lets you slip-edit or drag the
beginning of each track to the right to remove noises there.
You can cut out silent areas in each track to reduce leakage and background noise. Say you’ve recorded a real drum set. Look at the tomtom tracks. The tom mics pick up a lot of unwanted leakage from
the snare drum and cymbals. In the tom-tom tracks, you can cut out
the space between tom hits. This prevents the leakage and tightens
up the sound. Figure 13.16 shows some edited tom-tom tracks with
317
318
CHAPTER 13 Computer Recording
FIGURE 13.16
Edited tom-tom tracks with only the hits remaining.
only the hits remaining. You could use a noise gate instead of doing
this editing.
DUPLICATE MUSICAL PARTS
Suppose you have a perfect take of the background vocals in a chorus. Rather than recording more parts for the other choruses, you
could copy the vocals from one chorus and paste them at the measures where the chorus is repeated. Take care to align the vocal clips
in time with the other tracks.
REPLACE WRONG NOTES
Does the bass track have a few fluffed notes? Simply delete a wrong
note, copy an in-tune note, and paste it where the wrong note was.
Zoom-in to adjust the timing with precision.
MAKE SPECIAL EFFECTS
Editing can create unusual effects. For instance, copy a syllable in a
vocal track and paste it several times in a row to create stuttering.
Suppose you want to double a guitar that is in the left channel to
make it stereo. Copy the guitar track, paste it to another track, slide
Computer Recording CHAPTER 13
FIGURE 13.17
A crossfade between two songs.
the pasted guitar track to the right about 20 to 30 msec (which delays
the guitar signal), and pan the delayed guitar track to the right.
Adjust the track faders so that you hear the guitar spread evenly
between your monitor speakers.
FADE-OUT, FADE-IN
A fade-out at the end of a song is a slow drop in volume to silence.
DAW programs let you insert a fade-out at the end of a finished mix
(or on individual tracks). If you want to fade-in a synth-pad intro,
you can do that too. Fade-out and fade-in curves come in various
shapes, so use whatever sounds best.
CROSSFADE
This means, “fade-out of one song while fading into another.” Or
crossfade across an edit point to make it sound smoother. Figure
13.17 shows two songs overlapping. The first song fades out as the
other song fades in, and the two songs blend where they overlap.
FIX TIMING ERRORS
With most DAW programs you can slip each track forward or backward in time (left or right on the screen) independent of other
319
320
CHAPTER 13 Computer Recording
A
B
FIGURE 13.18
(A) Two musical parts out of sync. (B) Two parts in sync.
tracks. You also can move single notes in time. If a note in a bass
track comes in too late, highlight that note as a separate clip, and
slide it or nudge it to the left until it is on the beat. By visually aligning the start times (attacks) of notes in various tracks, you can make
a recorded band sound very tight.
Figure 13.18 shows two tracks that I wanted to synchronize. I set the
DAW to display only those two tracks. In Figure 13.18A, the notes in
the bottom track play later than the notes in the top track. In Figure
13.18B, the notes in the bottom track have been aligned with the
notes in the top track—they play in sync.
REMOVE OR REARRANGE SONG SECTIONS
You do this type of edit on a finished stereo mix, not necessarily on
individual tracks.
Suppose you want to shorten a song by removing a section. Highlight
the section you want to delete, cut it (Ctrl-X) and close up the space.
Zoom into the edit point and slide the beginning or end of the clips
there to make a smooth transition. Try to create a zero-crossing at
the edit point, so that the waveform crosses the 0 dB line there (as
in Figure 16.8). This method also works for editing mistakes out of
spoken-word recordings. Just be careful not to leave double breaths
during pauses.
Now suppose you want to remove a section of a song and put it
somewhere else in the song. If you think the bridge should come
Computer Recording CHAPTER 13
earlier, you can remove it from its current location and put it where
you want it.
1. Highlight the bridge waveform.
2. Cut it (Ctrl-X) and close up the space.
3. Split the song where you want the bridge to go.
4. Slide the right-side part of the song to the right to create a space.
5. Paste the bridge waveform in that space.
6. Slide the waveform segments to create a seamless edit.
CHANGE THE SONG ORDER WHILE MASTERING
When mastering a program for CD or MP3 format, you can select entire
songs and move them around to change the order they will play.
Those are just a few ways that editing can improve your mixes and
song arrangements.
Maintaining Audio Quality
Do your computer mixes sound harsh, grainy, or “digital”? Here
are some tips to make your in-the-box mixes sound as smooth
as analog.
Suppose you’re recording and mixing with a DAW set to 24-bit resolution. You think you’re doing everything right, but the final product
still sounds vaguely distorted. You record tracks near 0 dBFS but no
higher, and your meters don’t show any clipping. Still, your mixes
lack that smooth, effortless sound you used to get when recording
analog. The sound is harsh or fatiguing.
What’s going on? Chances are the signal is clipping but the meters
don’t indicate it. We’ll explain this phenomenon, and will suggest
ways to solve that problem. We’ll also give other tips on helping your
audio to sound less processed and more musical.
BACKGROUND
First we need to go over some digital audio basics. The A-to-D converter in your audio interface measures or samples the incoming
analog signal many thousand times a second and converts those
samples to binary numbers, which you record on your hard drive.
321
CHAPTER 13 Computer Recording
Intersample peak level
Sample level (meter reading)
0.6
0.4
0.2
Volts
322
0.0
Time
–0.2
–0.4
–0.6
FIGURE 13.19
Analog signal levels can exceed digital sample levels.
Figure 13.19 shows an analog sine wave that is sampled periodically.
The digital meters in your DAW show those sample levels, but not
necessarily the peak signal levels which appear between samples. In
your D-to-A converter, a lowpass filter (reconstruction filter) creates
occasional peaks between the samples that can be up to 3 dB higher
than the measured level at certain high frequencies. If your recorded
levels are near 0 dBFS, this can cause clipping (overs or overloads)
that the meters don’t display. An “over” is a segment of audio in
which three or more consecutive samples are at 0 dBFS (all bits on).
Similarly, some plug-ins can create larger peaks than those that went
in—without any increase in volume or audible change in the program. Even an EQ cut or rolloff (subtractive EQ) can increase the signal level by producing intersample peaks due to ringing at the cut
frequency.
With these facts in mind, let’s explore some ways to prevent harsh
digital sound.
USE METERS CORRECTLY
First, let’s review the concept of average levels and peak levels. As
shown in Figure 13.20, a musical signal changes in level (voltage)
continuously as it plays. Imagine a musical passage with a low-level
synth pad, but with high-level drum hits. The average level or volume of the passage is low, but the transient peak levels are high.
Computer Recording CHAPTER 13
Clipping level
Peak levels
Average levels
Average levels
Peak levels
Clipping level
FIGURE 13.20
Average levels and peak levels in a musical signal.
Peak levels may be up to 24 dB above average levels depending on
the type of signal. Percussive sounds have much higher peaks than
continuous sounds do (synth pads, organ, flute)—even if the two
signals have similar average levels.
The meters in your DAW can show signal levels in two modes: rms
and peak. The rms (or root-mean-square) readings correspond to the
average levels, while peak readings show the level of peaks or short
transients. The average or rms level indicates approximately how
loud the sound is, while the peak level shows how close the signal is
to clipping.
Since you don’t want to clip or distort the signal while recording or
mixing, use peak metering, not rms.
REDUCE RECORDING LEVELS
To prevent intersample clipping, reduce your recording levels by turning down the gain in your mic preamps. Record at about ⫺6 dBFS
maximum in peak meter mode. Make sure that the recording meters
are set to pre-fader.
Some DAWs, such as Cakewalk Sonar Producer, color the top 6 dB of
the meters red as an aid to setting levels correctly.
One benefit of lower recording levels is that you won’t overdrive your
mic preamp. The distortion of most analog gear increases as you
approach maximum gain. The mic preamp in your audio interface is
less likely to distort its analog signal at 6 dB below maximum than at
3 to 0 dB below maximum.
323
324
CHAPTER 13 Computer Recording
Another benefit of reduced levels is that it creates some headroom
for your plug-ins; 6 dB of headroom should eliminate any invisible
overs. Then you can set up your mix balances without having to
adjust levels every time you insert a plug-in. With lower track levels, you won’t hear distortion from plug-ins and your D/A converter
while you’re mixing.
Signals can clip during any gain boost, such as by a fader, compressor, or EQ. So keep recorded levels low (⫺6 dBFS max) so you don’t
have to change plug-in levels during a mix. Each track’s waveform
may appear small, but most DAWs let you zoom in vertically.
Remember, the recorded level on each track drives the plug-ins—
not the track fader, which comes after the plugs. If a track was
recorded at 0 dBFS maximum, use your DAW’s trim control (or insert
a ⫺6 dB trim plug-in) to reduce the signal level going into your
plugs. Do this before setting up a compressor or gate, because trim
affects their gain reduction.
Doesn’t a low recording level increase noise? Yes, but in practice
it’s not a problem. A 24-bit recording has a theoretical S/N ratio of
144 dB. So even if you record at ⫺6 dB max, you’re still far above the
noise floor and you won’t hear any increase in hiss.
Low noise is important because the more tracks in use, the higher
the noise. For instance adding 16 tracks together at equal fader settings will increase noise by about 12 dB. Still, the S/N is more than
you need to have a clean recording. So you can use that extra S/N
to create headroom by recording and mixing at lower levels. If your
peak recording level is ⫺6 dBFS, you create 6 dB of headroom. If your
peaks are ⫺10 dBFS, you create 10 dB of headroom.
An alternative is to get an oversampling meter plug-in which shows
the actual intersample signal levels. By modeling the D/A conversion
process, this type of meter reconstructs intersample peaks that regular meters miss. Examples include: Elemental Audio Inspector XL and
TL MasterMeter for ProTools, Sonoris Meter for SAW Studio, Sony
Oxford Limiter, TCE 6000 MD3 algorithm, TCE MD3 PowerCore
plug-in, and Voxengo Elephant.
If you just keep your levels down, you don’t have to worry about
intersample peaks.
Computer Recording CHAPTER 13
REDUCE LEVELS AT OTHER POINTS
Listed below are other spots in the signal path to set your level to
⫺6 dB maximum:
■
■
■
On any stem (group or submix).
At the output of each plug-in. Check for distortion (clipping)
in each plug-in, especially in an equalizer set to boost. Find the
plug’s output gain control and turn it down until clipping stops.
You might have several plugs in series, and you don’t want the
output of one plug to overload the input of another.
At the output of any upsampling plug-ins. These DSP processes
upsample the signal (increase its sample rate), do the effect,
then downsample the signal at the output. The downsampler
acts like a partial reconstruction filter. So even with the effect
turned off, the up and down sampling can create overs.
Level setting is important during mixdown, too. Set the maximum
peak level of the master output bus meters to ⫺3 to ⫺6 dBFS to allow
for signal peaks that don’t display on sample-reading meters. You can
make up the gain later during mastering, where you can boost the
overall level or peak-limit/normalize to make a hot CD or MP3 file.
Try to keep the master faders at or near 0 dB. If the master faders are
set low, you will turn up the channel faders to get a good mix level.
That results in high-level signals that can overload the mix bus. Still,
if a master-fader meter clips, it’s okay to lower the master fader a few
dB until the clipping stops.
Note that mix-bus overload isn’t a serious problem if the DAW uses
48- or 64-bit floating-point math in the mix bus, because float processing can pass signals above 0 dBFS without clipping. Still, it’s good
practice to control the levels hitting the mix bus. Summing calculations become less accurate when signals exceed full scale.
Consider not using makeup gain in compressors. Makeup gain can
cause the initial uncompressed transient to clip by boosting its level.
Raise the track fader instead.
If your mix is going to a mastering engineer, omit any stereo bus processing. Record the file at ⫺6 to ⫺3 dBFS max so that the mix doesn’t
overload the mastering engineer’s D/A converter. If you’re mastering
325
326
CHAPTER 13 Computer Recording
your own mixes (say, for customer approval), use an oversampling
meter that displays the actual reconstructed signal, or set the limiter
ceiling at ⫺3 dB to ⫺0.3 dB, not 0 dB or higher. This avoids making
files that sound distorted on your customer’s playback systems. Many
CD or MP3 players distort with samples above the clip level.
So far we looked at level reduction as a path to smoother sound. Now
let’s consider other methods to make digital audio easier to listen to.
WORK ON THE SOUND SOURCE
Sometimes a musical instrument or vocal sounds edgy, even when
recorded at low levels. The cause might be the source itself, the
microphone frequency response or mic placement. Here are some
ways to make the source sound better:
■
■
■
■
■
With EQ, consider using less high-frequency boost. Consider
cutting just a few dB around 3 to 7 kHz, which tends to reduce
harshness. Try a 4 kHz lowpass (high-cut) filter on distorted guitar amps to take the edge off.
Try multiband compression on voices that get raspy at high volume. Set the compressor so it kicks in above 3 kHz or so during
loud passages.
Many condenser mics have high-frequency peaks in their
response which can sound brittle or sibilant. Prevent that effect
by applying a high-frequency rolloff or shelf in your EQ. Better
yet, use ribbon mics or flat-response condenser mics. Many
ribbon microphones have reduced output at high frequencies
which can give a rounder tone.
Move the mic to a spot that sounds more mellow. For example,
mike a vocal partly off to the side if it sounds too edgy when
miked in front. Mike a trumpet off-axis. Mike a guitar amp near
the edge of the speaker cone.
Try using a tube mic preamp to warm up the sound. Not all
tube preamplifiers do that, but some have euphonic even-order
distortion that some people call added “warmth.”
OTHER TIPS
■
Dirty contacts can cause signal rectification and distortion.
Clean connectors with agents such as Caig Labs DeoxIT (www
.caig.com). Consider using high-quality cables.
Computer Recording CHAPTER 13
■
■
■
■
■
■
■
■
■
■
■
■
Use better converters.
Use audio interfaces designed for low jitter.
During mastering, add dither when truncating from 24- to 16bit. This eliminates quantization distortion at low signal levels.
Minimize the number of calculations that your DAW has to do.
Each calculation creates some error or distortion by increasing
the word length beyond 24 bits. Rather than boosting the gain
of a musical section and dropping it later, undo the changes
and apply only the correct amount of gain change.
Consider using volume envelopes (fader moves) rather than compression. Compression adds distortion because it changes the
waveform. Too-fast release times cause distortion in low-frequency
notes and also cause pumping. A fader-setting change on a series
of notes is less audible than a compressor working on each note.
Limiting adds distortion. When you peak-limit and normalize
a stereo mix to make a hot CD or MP3 file, try not to exceed 6
to 7 dB of peak reduction. Use even less limiting on mixes that
don’t have loud transients. Rather than normalizing the limited signal to 0 dBFS, normalize to ⫺3 dB to ⫺1 dB to allow for
intersample peaks.
Experiment with an analog tape plug-in to warm up a track
(e.g., Crane Song Phoenix). Or record to analog multitrack tape
and dump the tracks to your DAW for editing/mixing.
To reduce graininess in a reverb plug-in, use a high-density setting or use a convolution reverb.
For best quality, record at 24-bits and stay there throughout the
project. Then turn on dither and export to a 16-bit file that you will
master for a CD or MP3 file. If you’re sending 24-bit data files to a
mastering engineer, don’t turn on dither—let the engineer do it.
Normalization is a DAW process that raises the gain of the
entire program so that the highest peak reaches 0 dBFS (in peak
meter mode, not rms). Don’t normalize the program until final
mastering.
Use cables designed for digital audio, and use short cables. Use
internal sync on your A/D converter to reduce jitter.
Avoid A/D and D/A conversions. Once your signal is digital, try
to keep it that way. After converting the analog mic signal to digital in your audio interface, do all your processing in the computer
if possible, then burn a CD or MP3 file of the final product. The
only D/A conversion will be when the CD or MP3 file plays back.
327
328
CHAPTER 13 Computer Recording
Give all these suggestions a try—your music will sound better.
As we’ve seen, there is a whole world of computer music applications. Download the free trial versions to see what you like. Set
up your DAW system, read the software manual, and have fun creating music.
AUDIO FOR VIDEO
You can use your DAW to create a sound track for a video or film
production.
First we need to define the Society of Motion Picture and Television
Engineers (SMPTE) timecode, which is a standardized timecode signal for use in video productions. SMPTE timecode is something like
a digital tape counter, where the counter time is recorded as a signal
on tape or hard disk.
Pictures on a video screen are updated approximately 30 frames per
second, where a frame is a still picture. SMPTE timecode assigns a
unique number (address) to each video frame—8 digits that specify
hours:minutes:seconds:frames.
Suppose you want create an audio sound track for a video program.
First you copy the video onto your computer’s hard drive. When
you open the video file in your audio editing program, it shows up
in a window on your computer screen (Figure 13.21, top). While
watching the video, you create the sound track by editing audio and
music clips.
Here is the process:
■
■
■
First, you’re handed a DVD or sent a file which is a video
recording of the program you’re working on, plus location
audio. Copy the file(s) to your hard drive.
Launch your audio software, import the video file, and watch
the picture. Using soft synths, compose and record musical
parts related to the video scenes. Slide the audio clips so they
sync with the video.
Missing or poorly recorded dialog is replaced during a process
called looping or automatic dialog replacement (ADR). Have
actors watch the video and lip-sync their lines using the DAW.
Computer Recording CHAPTER 13
FIGURE 13.21
A video window and sound clips in a DAW screen.
■
■
Audition several sound effects from CD libraries, pick the ones
you like, and import them into tracks in the DAW.
Using slow motion or freeze frame, go through the video and
note where each sound effect should occur. Nudge audio clips
in time to align with video events.
Below is a more detailed procedure. I’m using these terms in the
steps below:
AUDIO EDITOR: Performs the audio edits and maybe composes
the music.
AUDIO MIXER: Mixes the end product for a feature film. There
might be several mixers: one for dialog, one for sound effects
and one for music.
1. During a video shoot, the audio from the location audio mixer
is recorded directly to camera.
2. The audio editor receives a video file with the location sound
track embedded (but extractable). In addition the audio editor
329
330
CHAPTER 13 Computer Recording
gets either a WAVE, OMF, or an AAF file of the audio to mix.
The audio track that is embedded in the video file is usually
a scratch mix to make sure the audio from the WAVE, OMF,
or AAF file is in sync with the video. You line them up by ear
(and eye).
3. In Hollywood film productions, the audio editor delivers to
the audio mixer either an OMF or AAF file that contains all the
audio edits in sync with the video. The audio in this file is in
multitrack format so the audio mixer has maximum flexibility
in mixing.
4. The audio mixer lines up the video and audio file at SMPTE
time 01:00:00:00, the traditional project start time, and starts
mixing. The video and audio in the computer are synchronized
by the audio interface at the resolution of the sample rate of
the mixer’s system (typically every 48,000th of a second).
5. After the audio mixer finishes the mix using the audio elements from the audio editor, he or she sends the final mixed
file (WAV or AIFF) to the video editor. The video editor marries
the mixed track to the video on a video editing system.
In smaller productions, the audio editor/mixer follows this procedure:
1. Receive a WAVE, OMF, or AAF file from the video editor.
2. Import location audio, narration, and sound effects. Compose
and record music. Sync these elements to picture and do a mix.
3. Export the mix (usually as a Broadcast WAV or AIFF file). Put a
timecode stamp (01:00:00:00) at the beginning of the file.
4. Send the mix to the video editor.
Thanks to Matthew Gellert of bonesteelfilms.com and David Schmidt
of acapellaaudio.com for their expert help with this section.
CHAPTER 14
Judging Sound Quality
Seat an engineer behind a mixing console and ask him to do a mix.
It sounds great. Then seat another engineer behind the same console
and again ask for a mix. It sounds terrible. What happened?
The difference lies mainly in their ears—their critical listening ability. Some engineers have a clear idea of what they want to hear and
how to get it. Others haven’t acquired the essential ability to recognize good sound. By knowing what to listen for, you can improve
your artistic judgments during recording and mixdown. You’re able
to hear errors in microphone placement, equalization, and so on,
and correct them.
To train your hearing, try to analyze recorded sound into its components—such as frequency response, noise, reverberation—and
concentrate on each one in turn. It’s easier to hear sonic flaws if you
focus on a single aspect of sound reproduction at a time. This chapter is a guide to help you do this.
CLASSICAL VERSUS POPULAR RECORDING
Classical and popular music have different standards of “good
sound.” One goal in recording classical music (and often folk music
or jazz) is to accurately reproduce the live performance. This is a
worthy aim because the sound of an orchestra in a good hall can be
quite beautiful. The music was composed and the instruments were
designed to sound best when heard live in a concert hall. The recording engineer, out of respect for the music, should always try to translate that sound to disc with as little technical intrusion as possible.
331
332
CHAPTER 14 Judging Sound Quality
By contrast, the accurate translation of sound to disc isn’t always the
goal in recording popular music. Although the aim may be to reproduce the original sound, the producer or engineer may also want to
play with that sound to create a new sonic experience, or to do some
of both.
In fact, the artistic manipulation of sounds through studio techniques has become an end in itself. Creating an interesting new
sound is as valid a goal as re-creating the original sound. There are
two games to play, each with its own measures of success.
If the aim of a recording is realism or accurate reproduction, the
recording is successful when it matches the live performance heard
in the best seat in the concert hall. The sound of musical instruments
is the standard by which such recordings are judged.
When the goal is to enhance the sound or produce special effects (as
in most pop-music recordings), the desired sonic effect is less defined.
The live sound of a pop group could be a reference, but pop-music
recordings generally sound better than live performances—recorded
vocals are clearer and less harsh, the bass is cleaner and tighter, and
so on. The sound of pop music reproduced over speakers has developed its own standards of quality apart from accurate reproduction.
GOOD SOUND IN A POP-MUSIC RECORDING
Currently, a good-sounding pop recording might be described as follows (there are always exceptions):
■
■
■
■
■
■
■
Well-mixed
Wide-range
Tonally balanced
Clean
Clear
Smooth
Spacious
It also has:
■
■
■
■
Presence
Sharp transients
Tight bass and drums
Wide and detailed stereo imaging
Judging Sound Quality CHAPTER 14
■
■
■
Wide but controlled dynamic range
Interesting sounds
Suitable production
The next sections explore each one of these qualities in detail so that
you know what to listen for. Assume that the monitor system is accurate, so that any colorations heard are in the recording and not in the
monitors.
A Good Mix
In a good mix, the loudness of instruments and vocals is in a pleasing balance. Everything can be clearly heard, yet nothing is obtrusive.
The most important instruments or voices are loudest; less important parts are in the background.
A successful mix goes unnoticed. When all the tracks are balanced
correctly, nothing sticks out and nothing is hidden. Of course, there’s
a wide latitude for musical interpretation and personal taste in making a mix. Dance mixes, for example, can be very severe sonically.
Sometimes you don’t want everything to be clearly heard. On rare
occasions you may want to mix in certain tracks very subtly for a
subconscious effect.
The mix must be appropriate for the style of music. For example, a
mix that’s right for loud rock music usually won’t work for a pop ballad. A rock mix typically has the drums way up front and the vocals
only slightly louder than the accompaniment. In contrast, a pop ballad has the vocals loudest, with the drums used just as “seasoning”
in the background.
Level changes during the mix should be subtle, or should make
sense. Otherwise, instruments jump out for a solo and fall back in
afterwards. Move faders slowly, or set them to preset positions during pauses in the music. Nothing sounds more amateurish than a
solo that starts too quietly and then comes up as it plays—you can
hear the engineer working the fader.
Wide Range
Wide range means extended low- and high-frequency response.
Cymbals should sound crisp and distinct, but not sizzly or harsh;
333
CHAPTER 14 Judging Sound Quality
kick drum and bass should sound deep, but not overwhelming or
muddy. Wide-range sound results from using high-quality microphones and adequate EQ.
You might want to combine “hi-fi” and “low-fi” sounds in a single
mix. The low-fi sounds generally cover a narrow frequency range and
might be distorted.
Good Tonal Balance
The overall tonal balance of a recording should be neither bassy nor
trebly; that is, the perceived spectrum shouldn’t emphasize low or
high frequencies. Low bass, mid-bass, midrange, upper midrange,
and highs should be heard in equal proportions (Figure 14.1).
Emphasis of any one frequency band over the other eventually
causes listening fatigue. Dance club mixes, however, are heavy on the
bass end to get the crowd moving.
Recorded tonal balance is inversely related to the frequency response
of the studio’s monitor system. If the monitors have a highfrequency rolloff, the engineer will compensate by boosting highs
in the recording to make the monitors sound correct. The result is a
bright recording.
LOUDNESS
334
BASS
Depth,
power
20
100
MIDBASS
MIDRANGE
UPPER
MIDRANGE
Body,
fullness
Fullness
of high
notes
Presence,
edge
500
2k
HIGHS
Brilliance,
sizzle
8k
FREQUENCY IN Hz
The subjective loudness of various frequency bands should be about
equal. The frequency divisions shown here are somewhat arbitrary.
FIGURE 14.1
Loudness versus frequency of a pop recording with good sound.
20k
Judging Sound Quality CHAPTER 14
Before doing a mix, play over the monitors some commercial recordings whose sound you admire, to become accustomed to a commercial spectral balance. After your mix is recorded, play it back and
alternately switch between your mix and a commercial recording.
This comparison indicates how well you matched a commercial spectral balance. An effective tool for this purpose is Harmonic Balancer
(www.har-bal.com). Of course, you may not care to duplicate what
others are doing.
In pop-music recordings, the tonal balance or timbre of instruments
doesn’t necessarily have to be natural. Still, many listeners want to
hear a realistic timbre from acoustic instruments, such as the guitar,
flute, sax, or piano. The reproduced timbre depends on microphone
frequency response, microphone placement, the musical instruments
themselves, and equalization.
Clean Sound
Clean means free of noise and distortion. Hiss, hum, and distortion
are inaudible in a good recording. Distortion in this case means distortion added by the recording process, not distortion already present in the sound of electric-guitar amps or Leslie speakers. There are
exceptions to this guideline; some popular recordings have noise or
distortion added intentionally.
Clean also means “not muddy” or free of low-frequency ringing and
leakage. A clean mix is one that is uncluttered or free of excess instrumentation. This is achieved by arranging the music so that similar
parts don’t overlap, and not too many instruments play at once in
the same frequency range. Usually, the fewer the instruments, the
cleaner the sound. Too many overdubs can muddy the mix.
Clarity
In a clear-sounding recording, instruments don’t crowd or mask each
other. They are separate and distinct. As with a clean sound, clarity
arises when instrumentation is sparse, or when instruments occupy
different areas of the frequency spectrum. For example, the bass provides low frequencies, keyboards might emphasize mid-bass, lead
guitar provides upper midrange, and cymbals fill in the highs.
335
336
CHAPTER 14 Judging Sound Quality
In addition, a clear recording has adequate reproduction of each
instrument’s harmonics; that is, the high-frequency response is not
rolled off.
Smoothness
Smooth means easy on the ears, not harsh, uncolored. Sibilant
sounds are clear but not piercing. A smooth, effortless sound allows
relaxation; a strained or irritating sound causes muscle tension in the
ears or body. Smoothness is a lack of sharp peaks or dips in the frequency response, as well as a lack of excessive boost in the midrange
or upper midrange. It is also low distortion, such as provided by a
24-bit recording.
Presence
Presence is the apparent sense of closeness of the instruments—a
feeling that they’re present in the listening room. Synonyms are clarity, detail, and punch.
Presence is achieved by close miking, overdubbing, and using microphones with a presence peak or emphasis around 5 kHz. Using
less reverb and effects can help. Upper midrange boost helps, too.
Most instruments have a frequency range that, if boosted, makes
the instrument stand out more clearly or become better defined.
Presence sometimes conflicts with smoothness because presence
often involves an upper midrange boost, while a smooth sound
is free of such emphasis. You have to find a tasteful compromise
between the two.
Spaciousness
When the sound is spacious or airy, there is a sense of air around the
instruments. Without air or ambience, instruments sound as if they
are isolated in stuffed closets. (Sometimes, though, this is the desired
effect.) You achieve spaciousness by adding reverb, recording instruments in stereo, using room mics, or miking farther away.
Sharp Transients
The attack of cymbals and drums generally should be sharp and
clear. A bass guitar and piano may or may not require sharp attacks,
depending on the song.
Judging Sound Quality CHAPTER 14
Tight Bass and Drums
The kick drum and bass guitar should “lock” together so that they
sound like a single instrument—a bass with a percussive attack. The
drummer and bassist should work out their parts together so they hit
accents simultaneously, if this is desired.
To further tighten the sound, damp the kick drum and record the
bass direct. Rap music, however, has its own sound—the kick drum
usually is undamped and boomy, sometimes with short reverb
added. Equalize the kick and bass in complementary ways so that
they don’t mask each other; for example:
Kick: Boost 60 to 80 Hz, cut 600 Hz, boost 3 kHz.
Bass: Cut 60 to 80 Hz, boost 600 Hz, cut above 3 kHz.
Wide and Detailed Stereo Imaging
Stereo means more than just left and right. Usually, tracks should be
panned to many points across the stereo stage between the monitor speakers. Some instruments should be hard-left or hard-right,
some should be in the center, others should be half-left or halfright (panned to 10 o’clock and 2 o’clock). Try to achieve a stereo
stage that is well balanced between left and right (Figure 14.2).
Instruments that occupy the same frequency range can be made
more distinct by panning them to opposite sides of center.
You may want some tracks to be unlocalized. Backup choruses and
strings should be spread out rather than appearing as point sources.
Stereo keyboard sounds can wander between speakers. A lead-guitar
solo can have a fat, spacious sound.
Left
speaker
Rhythm
guitar
Right
speaker
Background harmony,
Stereo keyboards
Floor
Tom
Low
rack
tom
Lead
vocal,
snare, kick,
bass
Stereo reverberation
FIGURE 14.2
An example of image placement between speakers.
Hi-hat
High
rack tom
Lead
guitar
337
338
CHAPTER 14 Judging Sound Quality
There should also be some front-to-back depth. Some instruments
should sound close or up front; others should sound farther away.
Use different miking distances or different amounts of reverb on various tracks.
If you want the stereo imaging to be realistic (for a jazz combo, for
example), the reproduced ensemble should simulate the spatial layout of the live ensemble. If you’re sitting in an audience listening to
a jazz quartet, you might hear drums on the left, piano on the right,
bass in the middle, and sax slightly right. The drums and piano aren’t
point sources, but are somewhat spread out. If spatial realism is the
goal, you should hear the same ensemble layout between your speakers. On some commercial CDs, the piano and drums are spread all
the way between speakers—an interesting effect, but unrealistic.
Pan-potted mono tracks often sound artificial in that each instrument sounds isolated in its own little space. It helps to add some stereo reverberation around the instruments to “glue” them together.
Often, TV mixes are heard in mono. Hard-panned signals sound
weak in mono relative to center-panned signals. So pan sound
sources to 3 and 9 o’clock, not hard-right and hard-left.
Wide but Controlled Dynamic Range
Dynamic range is the range of volume levels from softest to loudest.
A recording with a wide dynamic range becomes noticeably louder
and softer, adding excitement to the music. To achieve this, don’t
add too much compression (automatic volume control). An overly
compressed recording sounds squashed—crescendos and quiet interludes lose their impact, and the sound becomes fatiguing.
Vocals often need some compression or gain-riding because they
have more dynamic range than the instrumental backup. A vocalist may sing too loudly and blast the listener, or sing too softly and
become buried in the mix. A compressor can even out these extreme
level variations, keeping the vocals at a constant loudness. Bass guitar also can benefit from compression.
Interesting Sounds
The recorded sound may be too flat or neutral, lacking character or
color. In contrast, a recording with creative production has unique
Judging Sound Quality CHAPTER 14
musical instrument sounds, and typically uses effects. Some of these
are equalization, echo, reverberation, doubling, chorus, flanging,
compression, distortion, and stereo effects.
Making sounds interesting or colorful can conflict with accuracy or
fidelity. That’s okay, but you should know the trade-off.
Suitable Production
The way a recording sounds should imply the same message as the
musical style or lyrics. In other words, the sound should be appropriate for the particular tune being recorded.
For example, some rock music is rough and raw. The sound should
be, too. A clean, polished production doesn’t always work for highenergy rock ‘n’ roll. There might even be a lot of leakage or ambience
to suggest a garage studio or nightclub environment. The role of the
drums is important, so they should be loud in the mix. The toms
should ring, if that’s desired.
New Age, disco, rhythm and blues, contemporary Christian, or pop
music is slickly produced. The sound is usually tight, smooth, and
spacious. Folk music and acoustic jazz typically sound natural and
have little or no reverb. Music in the digital hardcore genre has lots
of distortion.
Actually, each style of music is not locked into a particular style of
production. You tailor the sound to complement the music of each
individual tune. Doing this may break some of the guidelines of
good sound, but that’s usually okay as long as the song is enhanced
by its sonic presentation.
GOOD SOUND IN A CLASSICAL-MUSIC RECORDING
As with pop music, classical music should sound clean, wide-range,
and tonally balanced. But because classical recordings are meant to
sound realistic—like a live performance—they also require good
acoustics, a natural balance, tonal accuracy, suitable perspective, and
accurate stereo imaging (see Chapter 18).
Good Acoustics
The acoustics of the concert hall or recital hall should be appropriate
for the style of music to be performed. Specifically, the reverberation
339
340
CHAPTER 14 Judging Sound Quality
time should be neither too short (dry) nor too long (cavernous). Too
short a reverberation time results in a recording without spaciousness or grandeur. Too long a reverberation time blurs notes together,
giving a muddy, washed-out effect. Ideal reverberation times are
around 1.2 seconds for chamber music or soloists, 1.5 seconds for
symphonic works, and 2 seconds for organ recitals. To get a rough
idea of the reverb time of a room, clap your hands once, loudly, and
count the seconds it takes for the reverb to fade to silence.
A Natural Balance
When a recording is well-balanced, the relative loudness of instruments is similar to that heard in an ideal seat in the audience area.
For example, violins aren’t too loud or soft compared to the rest
of the orchestra; harmonizing or contrapuntal melody lines are in
proportion.
Generally, the conductor, composer, and musicians balance the
music acoustically, and you capture that balance with your stereo
mic pair. But sometimes you need to mike certain instruments or
sections to enhance definition or balance. Then you mix all the mics.
In either case, consult the conductor for proper balances.
Tonal Accuracy
The reproduced timbre or tone quality should match that of live
instruments. Fundamentals and harmonics should be reproduced in
their original proportion.
Suitable Perspective
Perspective is the sense of distance of the performers from the
listener—how far away the stage sounds. Do the performers sound
like they’re eight rows in front of you, in your lap, or in another
room?
The style of music suggests a suitable perspective. Incisive, rhythmically motivated works (such as Stravinsky’s Rite of Spring) sound best
with closer miking; lush, romantic pieces (a Bruckner symphony) are
best served by more distant miking. The chosen perspective depends
on the taste of the producer.
Judging Sound Quality CHAPTER 14
Closely related to perspective is the amount of recorded ambience or
reverberation. A good miking distance yields a pleasing balance of
direct sound from the orchestra and ambience from the concert hall.
Accurate Imaging
Reproduced instruments should appear in the same relative locations
as they were in the live performance. Instruments in the center of the
ensemble should be heard in the center between the speakers; instruments at the left or right side of the ensemble should be heard from
the left or right speaker. Instruments halfway to one side should be
heard halfway off center, and so on. A large ensemble should spread
from speaker to speaker, while a quartet or soloist can have a narrower spread.
It’s important to sit equidistant from the speakers when judging stereo imaging, otherwise the images shift toward the side on which
you’re sitting. Sit as far from the speakers as they are spaced apart.
Then the speakers appear to be 60 degrees apart, which is about the
same angle an orchestra fills when viewed from the typical ideal seat
in the audience (tenth row center, for example).
The reproduced size of an instrument or instrumental section should
match its size in real life. A guitar should be a point source; a piano
or string section should have some stereo spread. Each instrument’s
location should be as clearly defined as it was heard from the ideal
seat in the concert hall.
Reproduced reverberation (concert-hall ambience) should surround
the listener, or at least it should spread evenly between the speakers.
Surround-sound technology is needed to make the recorded ambience surround the listener, although spaced-microphone recordings
have some of this effect. Accurate imaging is illustrated in Figure 14.3.
There should be a sense of stage depth: front-row instruments sound
closer than back-row instruments.
TRAINING YOUR HEARING
The critical process is easier if you focus on one aspect of sound
reproduction at a time. You might concentrate first on the tonal
341
342
CHAPTER 14 Judging Sound Quality
C
LC
C
RC
LC
R
L
Live ensemble
L
RC
R
Reproduced
Left speaker ensemble Right speaker
Reverberation approaches
mics from all directions
Reverberation approaches listener
from the front, and, with surround
systems, from all directions
FIGURE 14.3
With accurate imaging, the sound-source location and size, and the reverberant field, are reproduced
during playback.
balance—try to pinpoint what frequency ranges are being emphasized or slighted. Next listen to the mix, the clarity, and so on.
Soon you have a lengthy description of the sound quality of your
recording.
Developing an analytical ear is a continuing learning process. Train
your hearing by listening carefully to recordings—both good and
bad. Make a checklist of all the qualities mentioned in this chapter.
Compare your own recordings to live instruments and to commercial recordings. Check out the Golden-Ear ear training Web site at
www.moultonlabs.com/gold.htm and click on “Product.”
A pop-music record that excels in all the attributes of good sound
is The Sheffield Track Record (Sheffield Labs, Lab 20), engineered
and produced by Bill Schnee. In effect, it’s a course in state-of-the-art
sound—required listening for any recording engineer or producer.
Another record with brilliant production is The Nightfly by Donald
Fagen (Warner Brothers 23696-2), engineered by Roger Nichols,
Daniel Lazerus, and Elliot Scheiner, produced by Gary Katz, and
Judging Sound Quality CHAPTER 14
mastered by Bob Ludwig. This recording, and Steely Dan recordings
by Roger Nichols, sound razor sharp, very tight and clear, elegant,
and tasteful; and the music just pops out of the speakers.
The following listings are more examples of outstanding rock production, and set high standards:
Songs:
“I
Need Somebody,” by Bryan Adams; producer, Bob
Clearmountain
“The Power of Love,” by Huey Lewis & The News; producer, Huey
Lewis & The News
Albums:
90125, by Yes; producer, Trevor Horn
Synchronicity, by The Police; producer, Hugh Padgham and The
Police
Dark Side of the Moon, by Pink Floyd; producer, Alan Parsons
Thriller, by Michael Jackson; engineer, Bruce Swedien; producer,
Quincy Jones
Avalon, by Roxy Music; engineer, Bob Clearmountain; producer,
Roxy Music
Nevermind, by Nirvana; producer, Butch Vig
Come Away With Me, by Norah Jones; engineer, Jay Newland
Genius Loves Company, by Ray Charles; several engineers
Give, by The Bad Plus; engineer, Tchad Blake
Live in Paris and The Look of Love, by Diana Krall; engineer, Al
Schmitt
Smile, by Brian Wilson; engineer, Mark Linett.
Some other recordings of note are those by Supertramp, Pat Metheny
Group, Alison Krauss, The Wailin’ Jennys, Heart, and Radiohead; and
by producers George Martin and Andy Wallace.
Then there are the incredibly clean recordings of Tom Jung (with
DMP records) and George Massenberg. Some classical-music recordings with outstanding sound are on the Telarc, Delos, and Chesky
labels. You can learn a lot by emulating these superb recordings and
many others.
Once you’re making recordings that are competent technically—clean,
natural, and well-mixed—the next stage is to produce imaginative
343
344
CHAPTER 14 Judging Sound Quality
sounds. You’re in command; you can tailor the mix to sound any way
that pleases you or the band you’re recording. The supreme achievement is to produce a recording that is a sonic knockout, beautiful, or
thrilling.
TROUBLESHOOTING BAD SOUND
Now you know how to recognize good sound, but can you recognize
bad sound? Suppose you’re monitoring a recording in progress, or listening to a recording you’ve already made. Something doesn’t sound
right. How can you pinpoint what’s wrong, and how can you fix it?
The rest of this chapter includes step-by-step procedures to solve audiorelated problems. Read down the list of “bad sound” descriptions until
you find one matching what you hear. Then try the solutions until
your problem disappears. Only the most common symptoms and
cures are mentioned; console maintenance isn’t covered.
This troubleshooting guide is divided into four main sections:
1. Bad sound on all recordings (including those from other studios)
2. Bad sound on playback only (the mixer output sounds all
right)
3. Bad sound in a pop-music recording
4. Bad sound in a classical-music recording
Before you start, check for faulty cables and connectors. Also check
all control positions; rotate knobs, and flip switches to clean the
contacts, and clean connectors with DeoxIT from Caig Labs.
Bad Sound on All Recordings
If you have bad sound on all your recordings, including those from
other studios, follow this checklist to find the problem:
■
■
■
■
■
■
■
■
Upgrade your monitor system.
Adjust tweeter and woofer controls on speakers.
Adjust the relative gains of tweeter and woofer amplifiers in a
bi-amped system.
Relocate speakers.
Improve room acoustics.
Equalize the monitor system.
Try different speakers.
Upgrade the power amp and speaker cables.
Judging Sound Quality CHAPTER 14
■
Monitor at a moderate listening volume, such as 85 dBSPL. We
hear less bass and treble in a program if it’s monitored at a low
volume, and vice versa. If we hear too little bass due to monitoring at a low level, we might mix in too much bass.
Bad Sound on Playback Only
You might have bad sound on your playback only, but your mixer
output sounds okay. If DAT tape playback has glitches or drop-outs,
try these steps:
■
■
■
■
Clean the recorder with a dry cleaning tape.
Before recording, fast-forward the tape to the end and rewind it
to the top.
Use better tape.
Format videocassettes nonstop from start to finish.
If a hard-drive recording has glitches or drop-outs on playback, try
this:
■
■
Increase the latency setting in your recording software.
Follow the tips in Appendix B.
If a digital recording sounds distorted, these suggestions might help:
■
■
■
■
■
■
Keep the recording level as high as possible, but try not to
exceed ⫺6dBFS in peak meter mode, not rms mode.
Avoid clipping in effects plug-ins.
Record at a higher sampling rate or higher bit depth.
Avoid sampling-rate conversion.
Apply dither when going from a high bit depth to a lower bit
depth.
Try a higher quality audio interface.
Bad Sound in a Pop-Music Recording Session
Sometimes you have bad sound in a pop-music recording session.
MUDDINESS (LEAKAGE)
If the sound is muddy from excessive leakage, try the following:
■
■
Place the microphones closer to their sound sources.
Spread the instruments farther apart to reduce the level of the
leakage.
345
346
CHAPTER 14 Judging Sound Quality
■
■
■
■
■
■
■
■
Place the instruments closer together to reduce the delay of the
leakage.
Use directional microphones (such as cardioids).
Overdub the instruments.
Record the electric instruments direct.
Use baffles (goboes) between instruments.
Deaden the room acoustics (add absorptive material or flexible
panels).
Filter out frequencies above and below the spectral range of
each instrument. Be careful or you’ll change the sound of the
instrument.
Turn down the bass amp in the studio, or monitor the bass with
headphones instead.
MUDDINESS (EXCESSIVE REVERBERATION)
If the sound is muddy due to excessive reverberation, try these steps:
■
■
■
■
■
Reduce the effects-send levels or effects-return levels. Or don’t
use effects until you figure out what the real problem is.
Place the microphones closer to their sound sources.
Use directional microphones (such as cardioids).
Deaden the room acoustics.
Filter out frequencies below the fundamental frequency of each
instrument.
MUDDINESS (LACKS HIGHS)
If your sound is muddy and lacks highs, or has a dull or muffled
sound, try the following:
■
■
■
■
■
Use microphones with better high-frequency response, or use
condenser mics instead of dynamics.
Change the mic placement. Put the mic in a spot where there
are sufficient high frequencies. Keep the high-frequency sources
(such as cymbals) on-axis to the microphones.
Use small-diameter microphones, which generally have a flatter
response off-axis.
Boost the high-frequency equalization or cut slightly around
300 Hz.
Change musical instruments; replace guitar strings; replace
drum heads. (Ask the musicians first!)
Judging Sound Quality CHAPTER 14
■
■
■
■
■
Use an enhancer signal processor, but watch out for noise.
Use a direct box on the electric bass. Have the bassist play percussively or use a pick if the music requires it. When compressing the bass, use a long attack time to allow the note’s attack to
come through. (Some songs don’t require sharp bass attacks—
do whatever’s right for the song.)
Damp the kick drum with a pillow, folded towel, or blanket,
and mike it next to the center of the head near the beater. Use a
wooden or plastic beater if the song and the drummer allow it.
Don’t plug an electric guitar directly into a mic input. Use a
direct box or a high-impedance input.
Apply high-frequency boost after compression, not before.
MUDDINESS (LACKS CLARITY)
If your sound is muddy because it lacks clarity, try these steps:
■
■
■
■
■
■
Consider using fewer instruments in the musical arrangement.
Maybe turn down synth pads in the mix.
Equalize instruments differently so that their spectra don’t
overlap.
Try less reverberation.
Using equalizers, boost the presence range of instruments that
lack clarity. Or cut 1 to 2 dB around 300 Hz.
In a reverb unit, add about 45 to 100 msec of pre-delay.
Pan similar-sounding instruments to opposite sides.
DISTORTION
If you hear distortion when monitoring the mics in a pop-music
recording, try the following:
■
■
■
■
■
Increase input attenuation (reduce input gain), or plug in a pad
between the microphone and mic input.
Readjust gain-staging: Set faders and pots to their design centers
(shaded areas).
If you still hear distortion, switch in the pad built into the
microphone (if any).
Check connectors for stray wires and bad solder joints.
Unplug and plug-in connectors. Clean them with Caig Labs
DeoxIT or Pro Gold.
347
348
CHAPTER 14 Judging Sound Quality
TONAL IMBALANCE
If you have bad tonal balance—the sound is boomy, dull, or shrill,
for example—follow these steps:
■
■
■
■
■
■
■
■
■
Change musical instruments; change guitar strings; change
reeds, etc.
Change mic placement. If the sound is too bassy with a directional microphone, you may be getting proximity effect. Mike
farther away or roll off the excess bass.
Use the 3:1 rule of mic placement to avoid phase cancellations.
When you mix two or more mics to the same channel, the distance between mics should be at least three times the mic-tosource distance.
Try another microphone. If the proximity effect of a cardioid
mic is causing a bass boost, try an omnidirectional mic instead.
If you must place a microphone near a hard, reflective surface,
try a boundary microphone on the surface to prevent phase
cancellations.
If you’re recording a singer/guitarist, delay the vocal mic signal
by about 1 msec.
Change the equalization. Avoid excessive boost. Maybe cut
slightly around 300 Hz if the sound is muddy, or cut around
3 kHz if the sound is harsh.
Use equalizers with a broad bandwidth, rather than a narrow,
peaked response.
During mastering, try a spectrum analyzer/equalizer such as
Harmonic Balancer (www.har-bal.com).
LIFELESSNESS
If your pop-music recording has a lifeless sound and is unexciting,
these steps might help you solve it:
■
■
■
■
Work on the live sound of the instruments in the studio to
come up with unique effects.
Add effects: reverberation, echo, exciter, doubling, equalization,
etc.
Use and combine recording equipment in unusual ways.
Try overdubbing little vocal licks or synthesized sound effects.
If your sound seems lifeless due to dry or dead acoustics, try these:
Judging Sound Quality CHAPTER 14
■
■
■
■
■
If leakage isn’t a problem, put microphones far enough from
instruments to pick up wall reflections. If you don’t like the
sound this produces, try the next suggestion.
Add reverb or echo to dry tracks. (Not all tracks require reverberation. Also, some songs may need very little reverberation so
that they sound intimate.)
Use omnidirectional microphones.
Add hard, reflective surfaces in the studio, or record in a hardwalled room.
Allow a little leakage between microphones. Put mics far
enough from instruments to pick up off-mic sounds from other
instruments. Don’t overdo it, though, or the sound becomes
muddy and track separation becomes poor.
NOISE (HISS)
Sometimes your pop-music recording has extra noise on it. If your
sound has hiss, try these:
■
■
■
■
■
■
■
■
■
■
■
Check for noisy guitar amps or keyboards.
Switch out the pad built into the microphone (if any).
Reduce mixer input attenuation (increase input gain).
Use a more sensitive microphone.
Use an impedance-matching adapter (a low- to high-Z step-up
transformer) between microphones and phone-jack mic inputs.
Use a quieter microphone (one with low self-noise).
Increase the sound pressure level at the microphone by miking
closer. If you’re using PZMs, mount them on a large surface or
in a corner.
Apply any high-frequency boost during recording, rather than
during mixdown.
If possible, feed recorder tracks from mixer direct outs or insert
sends instead of group or bus outputs.
Use a lowpass filter (high-cut filter).
As a last resort, use a noise gate.
NOISE (RUMBLE)
If the noise is a low-frequency rumble, follow these steps:
■
Reduce air-conditioning noise or shut off the air conditioning
temporarily.
349
350
CHAPTER 14 Judging Sound Quality
■
■
■
Use a highpass filter (low-cut filter) that is set around 40 to
100 Hz.
Use microphones with limited low-frequency response.
See the following section Noise (Thumps).
NOISE (THUMPS)
■
■
■
■
■
■
Change the microphone position.
Change the musical instrument.
Use a highpass filter set around 40 to 100 Hz.
If the cause is mechanical vibration traveling up the mic stand,
put the mic in a shock-mount stand adapter, or place the mic
stand on some carpet padding. Try to use a microphone that is
less susceptible to mechanical vibration, such as an omnidirectional mic or a unidirectional mic with a good internal shock
mount.
Use a microphone with a limited low-frequency response.
If the cause is piano pedal thumps, also try working on the
pedal mechanism.
HUM
Hum is a subject in itself. See the tips on hum prevention near the
end of Chapter 4.
POP
Pops are explosive breath sounds in a vocalist’s microphone. If your
pop-music recording has pops, try these solutions:
■
■
■
■
■
■
■
Place the microphone above or to the side of the mouth.
Place a foam windscreen (pop filter) on the microphone.
Stretch a nylon stocking over a crochet hoop, and mount it on a
mic stand a few inches from the microphone (or use an equivalent commercial product).
Place the microphone farther from the vocalist.
Use a microphone with a built-in pop filter (ball grille).
Use an omnidirectional microphone, because it is likely to pop
less than a directional (cardioid) microphone.
Switch in a highpass filter (low-cut filter) set around 80 to
100 Hz.
Judging Sound Quality CHAPTER 14
SIBILANCE
Sibilance is an overemphasis of “s” and “sh” sounds. If you’re getting
sibilance on your pop-music recording, try these steps:
■
■
■
■
■
Use a de-esser signal processor or plug-in. Or use a multiband
compressor, and compress only the range from 3 to 10 kHz.
Place the microphone farther from the vocalist.
Place the microphone toward one side of the vocalist, or at nose
height, rather than directly in front.
Cut equalization in the range from 5 to 10 kHz.
Change to a duller sounding microphone such as a ribbon.
BAD MIX
Some instruments or voices are too loud or too quiet. To improve a
bad mix, try the following:
■
■
■
■
■
Change the mix. (Maybe change the mix engineer!)
Compress vocals or instruments that occasionally get buried.
Change the EQ on certain instruments to help them stand out.
During mixdown, continuously change the mix to highlight
certain instruments according to the demands of the music.
Change the musical arrangement so that different musical parts
don’t play at the same time. That is, consider having a call-andresponse arrangement (fill-in-the-holes) instead of everything
playing at once, all the time.
UNNATURAL DYNAMICS
When your pop-music recording has unnatural dynamics, loud
sounds don’t get loud enough. If this happens, try these steps:
■
■
■
Use less compression or limiting.
Avoid overall compression.
Use multiband compression on the stereo mix instead of wideband (full-range) compression.
ISOLATED SOUND
If some of the instruments on your recording sound too isolated, as
if they aren’t in the same room as the others, follow these steps:
■
In general, allow a little crosstalk between the left and right
channels. If tracks are totally isolated, it’s hard to achieve the
351
352
CHAPTER 14 Judging Sound Quality
■
■
■
■
■
■
■
illusion that all the instruments are playing in the same room at
the same time. You need some crosstalk or correlation between
channels. Some right-channel information should leak into the
left channel, and vice versa.
Place microphones farther from their sound sources to increase
leakage.
Use omnidirectional mics to increase leakage.
Use stereo reverberation or echo.
Pan effects returns to the channel opposite the channel of the
dry sound source.
Pan extreme left-and-right tracks slightly toward center.
Make the effects-send levels more similar for various tracks.
To give a lead-guitar solo a fat, spacious sound, use a stereo
chorus. Or send its signal through a delay unit, pan the direct
sound hard left, and pan the delayed sound hard right.
LACK OF DEPTH
If the mix lacks depth, try these steps:
■
■
■
Achieve depth by miking instruments at different distances.
Use varied amounts of reverberation on each instrument. The
higher the ratio of reverberant sound to direct sound, the more
distant the track sounds.
To make instruments sound closer, use a longer pre-delay setting in your reverb (maybe 40 to 100 msec). Use a shorter predelay (under 30 msec) to make instruments sound farther away.
Bad Sound in a Classical-Music Recording
Check the following procedures if you have problems recording classical music:
TOO DEAD
If the sound in your classical recording is too dead—there isn’t
enough ambience or reverberation—try these measures to solve the
problem:
■
■
■
Place the microphones farther from the performers.
Use omnidirectional microphones.
Record in a concert hall with better acoustics (longer reverberation time).
Judging Sound Quality CHAPTER 14
■
■
Turn up the hall mics (if used).
Add artificial reverberation.
TOO CLOSE
If the sound is too detailed, too close, or too edgy, follow these steps:
■
■
■
■
■
■
Place the microphones farther from the performers.
Place the microphones lower or on the floor (as with a boundary microphone).
Roll off the high frequencies.
Use mellow-sounding microphones (many ribbon mics have
this quality).
Turn up the hall mics (if used).
Increase the reverb-send level.
TOO DISTANT
If the sound is distant and there is too much reverberation, these
steps might help:
■
■
■
■
■
■
Place the microphones closer to the performers.
Use directional microphones (such as cardioids).
Record with a spaced pair of directional mics aiming straight
ahead.
Record in a concert hall that is less live (reverberant).
Turn down the hall mics (if used).
Decrease the reverb-send level.
STEREO SPREAD TOO NARROW OR TOO WIDE
If your classical-music recording has a narrow stereo spread, try these
steps:
■
■
■
Angle or space the main microphone pair farther apart.
If you’re doing mid-side stereo recording, turn up the side output of the stereo microphone.
Place the main microphone pair closer to the ensemble.
If the sound has excessive stereo spread (or “hole-in-the-middle”),
try the following:
■
■
Angle or space the main microphone pair closer together.
If you’re doing mid-side stereo recording, turn down the side
output of the stereo microphone.
353
354
CHAPTER 14 Judging Sound Quality
■
■
In spaced-pair recording, add a microphone midway between
the outer pair, and pan its signal to the center.
Place the microphones farther from the performers.
LACK OF DEPTH
Try the following to bring more depth into your classical-music
recording:
■
■
■
Use only a single pair of microphones out front. Avoid
multi-miking.
If you must use spot mics, keep their level low in the mix.
Add more artificial reverberation to the distant instruments
than to the close instruments.
BAD BALANCE
If your classical-music recording has bad balance, try the following:
■
■
■
Place the microphones higher or farther from the performer.
Ask the conductor or performers to change the instruments’
written dynamics. Be tactful!
Add spot microphones close to instruments or sections needing
reinforcement. Mix them in subtly with the main microphones’
signals.
MUDDY BASS
If your recording has a muddy bass sound, follow these steps:
■
■
■
■
■
Aim the bass-drum head at the microphones.
Put the microphone stands and bass-drum stand on resilient
isolation mounts (such as a carpet pad), or place the microphones in shock-mount stand adapters.
Roll off the low frequencies or use a highpass filter set around
40 to 80 Hz.
Use artificial reverb with a shorter decay time at low frequencies.
Record in a concert hall with less low-frequency reverberation.
RUMBLE
Sometimes your classical-music recording picks up rumble from air
conditioning, trucks, and other sources. Try the following to clear
this up:
Judging Sound Quality CHAPTER 14
■
■
■
■
■
■
Check the hall for background rumble problems.
Temporarily shut off the air conditioning.
Record in a quieter location.
Use a highpass filter set around 40 to 80 Hz.
Use microphones with limited low-frequency response.
Mike closer and add artificial reverb.
DISTORTION
If your classical-music recording has distortion, try the following:
■
■
■
■
■
Switch in the pads built into the microphones (if any).
Increase the mixer input attenuation (turn down the input
trim).
Check connectors for stray wires or bad solder joints.
Avoid sample-rate conversion.
Apply dithering when going from 24 to 16 bits.
BAD TONAL BALANCE
Bad tonal balance expresses itself in a sound that is too dull, too
bright, or colored. If your recording has this problem, follow these
steps:
■
■
■
■
■
■
Change the microphones. Generally, use flat-response microphones with minimal off-axis coloration.
Follow the 3:1 rule mentioned in Chapter 7.
If a microphone must be placed near a hard, reflective surface,
use a boundary microphone on the surface to prevent phase
cancellations between direct and reflected sounds.
Adjust EQ. Try a spectrum analyzer/equalizer such as Harmonic
Balancer (www.har-bal.com).
Place the mics at a reasonable distance from the ensemble (tooclose miking sounds shrill).
Avoid mic positions that pick up standing waves or room
modes. Experiment with small changes in mic position.
This chapter described a set of standards for good sound quality in
both popular and classical music recordings. These standards are
somewhat arbitrary, but the engineer and producer need guidelines
to judge the effectiveness of the recording. The next time you hear
something you don’t like in a recording, the lists in this chapter will
help you define the problem and find a solution.
355
This page intentionally left blank
CHAPTER 15
Session Procedures,
Mastering, and CD Burning
“We’re rolling. Take One.” These words begin the recording session.
It can be an exhilarating or an exasperating experience, depending
on how smoothly you run it.
The musicians need an engineer who works quickly yet carefully.
Otherwise, they may lose their creative inspiration while waiting for the
engineer to get it together. And the client, paying by the hour, wastes
money unless the engineer has prepared for the session in advance.
This chapter describes how to conduct a multitrack recording session. These procedures should help you keep track of things and run
the session efficiently.
There are some spontaneous sessions, especially in home studios, that
just “grow organically” without advance planning. The instrumentation isn’t known until the song is done! You just try out different
musical ideas and instruments until you find a pleasing combination.
In this way, a band that has its own recording gear can afford to take
the time to find out what works musically before going into a professional studio. In addition, if the band is recording itself where it
practices, the microphone setup and some of the console settings can
be more or less permanent. This chapter, however, describes procedures usually followed at professional studios, where time is money.
PREPRODUCTION
Long before the session starts, you’re involved in preproduction—
planning what you’re going to do at the session, in terms of overdubbing, track assignments, instrument layout, and mic selection.
357
358
CHAPTER 15 Session Procedures, Mastering, and CD Burning
Instrumentation
The first step is to find out from the producer or the band what the
instrumentation will be and how many tracks will be needed. Make
a list of the instruments and vocals that will be used in each song.
Include such details as the number of tom toms, whether acoustic or
electric guitars will be used, and so on.
Recording Order
Next, decide which of these instruments will be recorded at the same
time and which will be overdubbed one at a time. It’s common to
record the instruments in the following order, but there are always
exceptions:
1. Loud rhythm instruments—bass, drums, electric guitar, electric
keyboards
2. Quiet rhythm instruments—acoustic guitar, piano
3. Lead vocal and doubled lead vocal (if desired)
4. Backup vocals (in stereo)
5. Overdubs—solos, percussion, synthesizer, sound effects
6. Sweetening—horns, strings
The lead vocalist usually sings a guide vocal or scratch vocal along
with the rhythm section so that the musicians can get a feel for the
tune and keep track of where they are in the song. You record the
vocalist’s performance but probably you will re-record it later. That
eliminates leakage and lets you focus on the lead vocal.
In a MIDI studio, a typical order might be:
■
■
■
■
■
■
Drum machine or soft-synth drum set (playing programmed
patterns)
Synthesizer bass sound
Synthesizer chords
Synth melody
Synth solos, extra parts
Vocals and miked solos
Track Assignments
Now you can plan your track assignments. Decide what instruments
will go on which tracks of the multitrack recorder. The producer may
have a fixed plan already.
Session Procedures, Mastering, and CD Burning CHAPTER 15
What if you have more instruments than tracks? Decide what groups
of instruments to put on each track. In a 4-track recording, for example, you might put guitars on track 1, bass and drums on track 2,
vocals on track 3, and keyboards on track 4.
Remember that when several instruments are assigned to the same
track, you can’t separate their images in the stereo stage. That is, you
can’t pan them to different positions—all the instruments on one
track sound as if they’re occupying the same point in space. For this
reason, you may want to do a stereo mix of the rhythm section on
tracks 1 and 2, for instance, and then overdub vocals and solos on
tracks 3 and 4.
It’s possible to overdub more than four parts on a 4-track recorder.
To do this, bounce (mix) several tracks onto one, then record new
parts over the original tracks. Your recorder manual describes this
procedure.
If you have many tracks available, leave several tracks open for experimentation. For example, you can record several takes of a vocal part
using a separate track for each take, so that no take is lost. Then combine the best parts of each take into a single final performance on
one track. Most recorder–mixers let you do these extra takes on virtual tracks.
It’s also a good idea to record the monitor mix on one or two
unused tracks. The recorded monitor mix can be used as a cue mix
for overdubs, or to make a recording for the client to take home and
evaluate.
Session Sheet
Once you know what you’re going to record and when, you can fill
out a session sheet (Figure 15.1). This simple document is adequate
for home studios. “OD” indicates an overdub. Note the recordercounter time for each take, and circle the best take.
Production Schedule
In a professional recording studio, the planned sequence of recording basic tracks and overdubs is listed on a production schedule
(Figure 15.2).
359
SONG: Escape to Air Island
TRACK
1
2
3
4
5
6
7
8
INSTRUMENT
BASS
KICK
DRUMS
LEAD VOC OD
HARM. VOC OD
LEAD GUIT OD
KEYS L
KEYS R
MICROPHONE
DIRECT
AKG D-112
CROWN GLM-100
STUDIO PROJECTS B1
STUDIO PROJECTS B1
SHURE SM57
DIRECT
DIRECT
TAKE 1 03:21 - 06:18 FS
2 06:25 - 09:24 INC
3 10:01 - 13:02
FIGURE 15.1
A session sheet for a home studio.
WEST WIND STUDIOS
PRODUCTION SCHEDULE
Artist: Steve Mills
Album: Long Distance Music
Producer: B. Brauning
Engineer: D. Scriven
Date: 1-17-09
Project files: c:\cakewalk projects\mills\
mr_potato_head.cwp and sambatina.cwp
Audio data folders: d:\mills\mr_potato_head\
and
\sambatina\
1. Record Mr. Potato Head
Instrumentation: Electric bass, drums, electric rhythm guitar, electric
lead guitar, acoustic piano, sax, lead vocal.
Comments: Record rhythm section with scratch vocal. OD sax, piano,
lead vocal. Double lead guitar in stereo.
2. Record Sambatina
Instrumentation: Electric bass, drums, acoustic guitar, percussion, synth
Comments: Record rhythm section with scratch acoustic guitar.
Record synth MIDI. OD acoustic guitar, percussion and synth.
3. Mix Mr. Potato Head
Comments: Add 40msec echo to toms. Increase reverb on sax solo.
4. Mix Sambatina
Comments: Add flanger to bass on intro only. Do automated flange
on percussion.
FIGURE 15.2
A production schedule.
Session Procedures, Mastering, and CD Burning CHAPTER 15
WEST WIND STUDIOS
TRACK SHEET
Artist: S. Mills
Producer: B. Brauning
Date: 1-17-09
Take # 1
Take # 2 NC
Take # 3 LFS
Take # 4 FS
Take # 5
Album: Long Distance Music
Engineer: D. Scriven
Song title: Dig Up Nebraska
Start 00:31
Start 03:54
Start 07:21
Start 09:01
Start 09:25
Stop 03:31
Stop 06:49
Stop 08:38
Stop 09:12
Stop 12:27
Track 1: elec bass (Fender Precision DI)
Track 2: kick
Track 3: snare
Track 4: rack tom
Track 5: floor tom
Track 6: overhead L
Track 7: overhead R
Track 8: rhythm gtr L (Martin HD-28)
Track 9: rhythm gtr R (Taylor 814ce)
Track 10: lead gtr (Parker PM-20PRO, Roland Cube 60)
Track 11: keys L (Prophet ‘08 DI)
Track 12: keys R (Prophet ’08 DI)
Track 13: scratch lead vocal
Track 14: lead vocal
Track 15: background vocals 1
Track 16: background vocals 2
FIGURE 15.3
A track sheet (multitrack log).
Track Sheet
Another document used in a pro studio is the track sheet or multitrack log (Figure 15.3). Write down which instrument or vocal goes
on which track. The track sheet also has blanks for other information such as take numbers. If you’re using a DAW, you can enter this
information by typing on-screen.
Microphone Input List
Make up a microphone input list similar to that seen in Table 15.1.
Later you will place this list by the mic snake box and by the mixing
console.
361
362
CHAPTER 15 Session Procedures, Mastering, and CD Burning
Table 15.1
A Microphone Input List
Input
Instrument
Microphone
1
Bass
Direct
2
Kick
EV N/D868
3
Snare
AKG C451
4
Overhead L
Shure SM81
5
Overhead R
Shure SM81
6
High toms
Sennheiser MD421-II
7
Floor tom
Sennheiser MD421
8
Electric lead guitar
Shure SM57
9
Electric lead guitar
Shure SM57
10
Piano treble
Crown PZM-6D
11
Piano bass
Crown PZM-6D
12
Scratch vocal
Beyer M88
Be flexible in your microphone choices—you may need to experiment with various mics during the session to find one giving the best
sound with the least console equalization. During lead-guitar overdubs, for example, you can set up a direct box, three close-up microphones, and one distant microphone—then find a combination that
sounds best.
Find out what sound the producer wants—a “tight” sound; a “loose,
live” sound; an accurate, realistic sound. Ask to hear recordings having the kind of sound the producer desires. Try to figure out what
techniques were used to create those sounds, and plan your mic
techniques and effects accordingly. Tips on choosing a microphone
are given in Chapter 6.
Instrument Layout Chart
Work out an instrument layout chart, indicating where each instrument will be located in the studio, and where baffles and isolation
booths will be used (if any). In planning the layout, make sure that
all the musicians can see each other and are close enough together to
play as an ensemble. Often a circular arrangement works well.
Session Procedures, Mastering, and CD Burning CHAPTER 15
Keep all these documents in a single folder or notebook labeled with
the band’s name and recording date. Also include contact information, time sheets, invoices, and all the correspondence about the
project.
SETTING UP THE STUDIO
About an hour before the session starts, clean up the studio to promote a professional atmosphere. Lay down rugs and place AC power
boxes according to your layout chart.
Now position the baffles (if any). Put out chairs, stools, and music
stands according to the layout. Run a headphone extension cable
from each artist’s location to the headphone junction box in the
studio.
Place mic stands approximately where they will be used. Wrap one
end of a microphone cable around each microphone-stand boom,
leaving a few extra coils of cable near the mic-stand base to allow
slack for moving. Run the rest of the cable back to the mic input
panel or snake box. Plug each cable into the appropriate wall-panel
or snake-box input according to your mic input list.
Some engineers prefer to run cables in reverse order, connecting to
the input panel first and running the cable out to the microphone
stand. That procedure leaves less of a confusing tangle at the input
panel where connections might be changed.
Now set up the microphones. Check each mic to make sure its
switches are in the desired positions. Put the mics in their stand
adapters, connect the cables, and balance the weight of each boom
against the microphone.
Finally, connect the musicians’ headphones for cueing. Set up a spare
cue line and microphone for last-minute changes.
SETTING UP THE CONTROL ROOM
Having prepared the studio, run through the Session Preparation
section of Chapter 12 to make sure the control room is ready for the
session. Then turn up the monitor system. Carefully bring up each
fader one at a time and listen to each microphone. You should hear
normal studio noise. If you hear any problems such as dead or noisy
363
364
CHAPTER 15 Session Procedures, Mastering, and CD Burning
microphones, hum, bad cables, or faulty power supplies, correct
them before the session.
Verify the mic input list. Have an assistant scratch each mic grille
with a fingernail and identify the instrument the microphone is
intended to pick up. If you have no assistant, listen on headphones
as you scratch the grilles.
Check all the cue headphones by playing a tone or music through
them and listening while wiggling each cable.
SESSION OVERVIEW
This is the typical sequence of events:
1. For efficiency, record the basic rhythm tracks for several songs
on the first session.
2. Do the overdubs for all the songs in a dubbing session.
3. Mix all the tunes in a mixdown session.
4. Edit the tunes and master the album.
After the musicians arrive, allow them 1/2 hour to 1 hour free set
up time for seating, tuning, and mic placement. Show them where
to sit, and work out new seating arrangements if necessary to make
them more comfortable.
Once the instruments are set up, listen to their live sound in the
studio and do what you can to improve it. A dull-sounding guitar
may need new strings, a noisy guitar amp may need new tubes, the
drums may need tuning, and so on. Adjust the studio lighting for the
desired mood.
RECORDING
Before you start recording, you might want to make connections to
record the monitor mix. This recording is for the producer to take
home to evaluate the performance.
Many DAW programs have session templates: a group of tracks with
EQ, compression and effects plug-ins already in place. You simply
load in the template and you’re ready to go. You can create your own
session templates, such as “16 tracks with aux 1 and 2” or “drum set.”
Session Procedures, Mastering, and CD Burning CHAPTER 15
Some DAWs let you create a track template, which is a format for
a single track. You might have a vocal track template with the
input channel, reverb, EQ, and compression already set up. Each
time you’re about to record another vocal, import the vocal track
template.
Follow the mixer recording procedures described in Chapter 12. Set
recording levels, then set cue mixes for the musicians’ headphones.
The monitor mix affects only what is heard, not what is recorded.
When you’re ready to record the tune, briefly play a metronome to
the group at the desired tempo, or play a click track (an electronic
metronome) through the cue system. Or just let the drummer set the
tempo with stick clicks.
Start recording. Note the recorder counter time. Hit the slate button
(if any) and announce the name of the tune and the take number.
Then the group leader or the drummer counts off the beat, and the
group starts playing.
The producer listens to the musical performance while the engineer watches levels and listens for audio problems. As the song progresses, you may need to make small level adjustments.
The assistant engineer (if any) runs the multitrack recorder and keeps
track of the takes on the track sheet, noting the name of the tune, the
take number, and whether the take was complete (Figure 15.3). Use
a code to indicate whether the take was a false start, long false start,
nearly completed, a “keeper,” and so on.
While the song is in progress, don’t use the solo function, because
the abrupt monitoring change may disturb the producer. The producer should stop the performance if a major flub (mistake) occurs
but should let the minor ones pass.
At the end of the song, the musicians should be silent for several
seconds after the last note. Or, if the song ends in a fade-out, the
musicians should continue playing for about 30 seconds so there is
enough material for a fade-out during mixdown.
After the tune is done, you can either play it back or go on to a second
take. If you connected your multitrack recorder to the insert jacks, use
the faders to set a rough mix with EQ and effects. The musicians will
365
366
CHAPTER 15 Session Procedures, Mastering, and CD Burning
catch their flubbed notes during playback; you just listen for audio
quality.
Now record other takes or tunes. Pick the best takes. Punch-in to correct errors on each best take, or fix errors with editing.
To protect your hearing and to prevent fatigue, try to limit tracking
sessions to four hours or less. Take breaks to give your ears and body
a rest.
Relating to the Musicians
During a session, the engineer needs not only technical skills, but
also people skills. It’s important to respect artistic personalities and
to keep the creative energy flowing.
First, learn the names of the band members and refer to them by
name during the session.
Musicians are often nervous at the beginning of a session. Having
a sense of humor and exuding confidence helps to put the artists at
ease. Talk about the band’s music, origin, or instruments. Be sure
that the artists are comfortable—adjust the layout and lighting, offer
beverages and snacks. Tell new clients that mistakes are normal and
are easy to fix. Ask band members, “Is the mix okay in your headphones? Is the volume all right?” “Does your recorded instrument
sound okay?”
Don’t badmouth other musicians, and respect the privacy of your clients. Each musician you record should feel confident that you’re not
telling others about their mistakes.
When you want to try something that could enhance a song, don’t
just do it. Ask first. Say, “What would you think about doubling the
guitar in stereo?” or “I want your input on this.”
During a playback, try not to point out the errors. The musicians
will hear them and will try to do better on the next take. Don’t say,
“That sucked.” Say “It’s almost there—do you want to punch in a
few spots?” Or say “That was good, but how about another take?”
The goal is to make the best record you can, not to make clients feel
bad about themselves. Give a musician time to practice during overdubs, too.
Session Procedures, Mastering, and CD Burning CHAPTER 15
If the band members are getting tired or are having trouble getting
through a particular song, you might say “Let’s take a break” or “Let’s
try another song and come back to this one later.”
Finally, if a piece of gear breaks or your software develops a glitch, try
to work around it quietly without making it obvious. Musicians need
an engineer who handles technical problems in a professional way.
OVERDUBBING
After recording the basic or rhythm tracks for all the tunes, add overdubs. A musician listens to recorded tracks over headphones and
records a new part on an open track. See the Chapter 12 sections
Overdubbing and Composite Tracks.
You might have the musician play in the control room while overdubbing. You can patch a synth or electric guitar into the console
through a direct box, and feed the direct signal to a guitar amp in
the studio via a cue line. Pick up the amp with a microphone, and
record and monitor the mic signal.
Do any drum overdubs right after the rhythm session because the
microphones are already set up, and the overdubbed sound will
match the sound of the original drum track.
BREAKING DOWN
When the session is over, tear down the microphones, mic stands,
and cables. Put the microphones back in their protective boxes and
bags. Wind the mic cables onto a power-cable spool, connecting one
cable to the next. Wipe off the cables with a damp rag if necessary.
Some engineers hang each cable in big loops on each mic stand.
Others wrap the cable “lasso style” with every other loop reversed.
You learn this on the job.
Put the labeled recording and session documents in a labeled box
or file folder. Normally the studio keeps the multitrack recording for
possible future work unless the group wants to buy or rent it. Backup
the multitrack recordings (audio files) and session files (track setups
for each song) to another medium, such as an external hard drive,
CD-R, or DVD-R.
367
368
CHAPTER 15 Session Procedures, Mastering, and CD Burning
If you’re using a hardware mixing console, log the console and
effects settings by writing them in the track sheet or reading them
slowly into a portable recorder. At a future session you can play back
the recording and reset the console the way it was for the original
session. Automated consoles can store and recall the console’s control adjustments. DAWs remember both the mixer settings and the
plug-in settings.
MIXDOWN
After all the parts are recorded, you’re ready for mixdown. Follow
the mixdown procedures in Chapter 12 and repeat for all the song
mixes. Again, be sure to backup your session files (the mix and edit
settings) and audio files (track audio recordings and mixes.)
MASTERING
After recording the mixes, you can burn a CD of the mixes as they
are, or you can master an album or demo of those mixes. Mastering
is the last creative step before burning the final CD used for duplication or replication. In mastering you edit out noises or false starts
before the beginning of each song, put the songs in the desired
order, insert a few seconds of silence between songs, and make each
song the same loudness and the same tonal balance. The goal is a
consistent sound from track to track, so that everything flows better
and the album sounds unified. You also might try to make the CD as
loud or “hot” as possible (without destroying its sound quality).
At this point, you have three choices:
1. If the CD will be just a reference, not a demo or album, you
can burn a CD of the unedited mixes and stop there.
2. If the CD will be a demo or album, burn a CD of the unedited
mixes, then send it to a mastering engineer for mastering.
3. Or you can edit and master the mixes yourself, then burn a
CD. It will be the CD pre-master for duplication or replication.
“Duplication” is making copies with a CD-R burner, while
“replication” is stamping CDs from a glass master (the professional standard).
Let’s look at each option above.
Session Procedures, Mastering, and CD Burning CHAPTER 15
Burning a Reference CD
You can use the CD-burning software that came with your computer
CD burner, or use other software. Some examples of CD-burning
programs are Roxio’s Easy Media Creator and Record Now Media
Lab, Nero, and Stomp’s Click ‘N Burn. Some DAW software includes
a CD-burning application.
Before you start, note that most CD burning software puts 2 seconds
of silence between songs. If you want to have extra seconds of silence
at the end of a song, record the silence as part of the song’s WAVE
file. You might want to normalize each song’s WAVE file. This raises
the level of the song so that the highest peak in the song reaches
maximum level: 0 dBFS or whatever level you specify. Normalizing
doesn’t make the tracks the same loudness, because loudness
depends on the average signal level, not the peak level.
Here’s a CD-burning procedure:
1. Start the CD-R burning software.
2. Select which WAVE files you want to put on the CD-R, drag
them to the playlist, and arrange them in the desired order.
The total playing time must be less than the CD-R length (74
or 80 minutes).
3. Choose “On the fly” or “Disc image.” When recording on-thefly, the computer grabs the sound files from random locations
on hard disk and puts them in order as the CD-R does a burn.
When recording a disk image, the computer rewrites the sound
files to a single, contiguous space on your hard disk and puts
them in order. Then you copy the disk image from hard disk
to CD-R. Disc image is less likely to produce CD errors than
On-the-fly.
4. Set the burn speed. Some CD recorders and blank CDs
can record up to 52 times normal speed. Normal speed is
172 kBps (kilobytes per second; Bytes is capital B), double
speed is 344 kBps, etc. A 52⫻ recorder can burn a 74-minute
CD in less than 2 minutes. Some CD burners automatically
select the optimum speed based on what the blank CD-R can
handle. High speeds don’t degrade sound quality seriously,
but they tend to increase errors—which the CD player may
or may not correct accurately. The recommended speed to
369
370
CHAPTER 15 Session Procedures, Mastering, and CD Burning
prevent errors is 2⫻ to 4⫻ when making a final master CD to
be sent out for replication. But some CD recorders and blank
CD-Rs have fewer errors at higher speeds.
5. Set the software to Disc-at-Once Mode. Start recording the
CD-R. The WAVE files will transfer in order to the CD-R disc.
To prevent glitches, don’t multitask while the CD is recording.
CD-R discs can’t be erased and used over again, so try to make
everything right before you burn a disc. You could do a practice burn on a CD-RW.
6. As soon as the recording is done, the display will indicate that
the table of contents is being written. Eventually, the system
will beep and eject the disc. To prevent error-causing fingerprints, be sure to handle the disc only by the edges. Pop the
disc in an audio CD player, press Play, and check that all the
tracks play correctly.
Sending Out Your CD for Mastering
You might prefer to send your CD of mixes to a good mastering engineer. This person can listen to your program with fresh ears, then
suggest processing for your album that will make it sound more
commercial. She is likely to have a better monitor system and better equipment than yours. They have heard hundreds of recording
projects done by others, and know how to make your CD sonically
competitive.
If you plan to have your program mastered outside, don’t apply
any signal processing to your finished mixes such as editing, level
changes, compression, normalization, fades, or EQ. Let the mastering house do it with their better equipment and software. Also, leave
some headroom by recording the finished mixes at about ⫺3 dBFS
maximum in peak meter mode.
Deliver your mixes in the highest possible resolution that the mastering engineer can accept, such as 96 to 192 kHz and 24 to 32 bits
on a data CD (ISO 9660 format). However, a 24-bit/44.1 kHz format
is usually adequate. Copy the mixes’ WAVE files to CD, rather than
creating CD audio tracks (.cda files). The resulting data CD can
be read by a CD-ROM but can’t be played on a CD player. If your
CD-burning program doesn’t support data CDs, either get one that
does, or create an audio or music CD at 16-bits/44.1 kHz.
Session Procedures, Mastering, and CD Burning CHAPTER 15
Also send the mastering engineer the sequence of songs and any
suggestions on how you would like certain songs to be improved or
edited. The mastering engineer might suggest that you need to re-do
some mixes.
Mastering Your Own Album
You can master an album with your multitrack recording software,
or with a mastering program such as Steinberg Nuendo, Sony Sound
Forge, BIAS Peak Pro, Sonalksis Mastering Suite, 1 K Multimedia
T-Racks 24, or Magix Samplitude.
First, discuss the order of the songs on the recording. The first track
on the album should be a strong, accessible, up-tempo piece. Change
keys from song to song. The last track should be as good as or better
than the first to leave a good final impression.
Mastering engineer Bob Katz offers this advice for song sequencing:
Create the album in sets of one to three songs of the same tempo.
Make sure the songs flow from one to the next. Here’s a suggested
song order:
1.
2.
3.
4.
5.
An up-tempo, exciting song that hooks the listener.
After a short space, an up-tempo or mid-tempo song.
After three or four songs, slow down the tempo.
Reach a climax near the end of the album.
The last song should sound relaxed and intimate, perhaps
using fewer members of the band.
You can drag the WAVE files of your song mixes into iTunes and
experiment with various sequences.
To get your mastered songs onto a CD, you need CD-burning
software, either consumer or pro. With a consumer CD-burning
program—such as Roxio’s Easy Media Creator or Record Now Media
Lab, Stomp’s Click ‘N Burn, or Nero—you drag the WAVE files of
each of your song mixes into a playlist, then burn the CD.
With a professional program, you create one long WAVE file of all the
song mixes. You can adjust the timing between each pair of songs.
You specify when each song starts, and the software creates a table
of contents on the CD made of song-start IDs. This feature is called
371
372
CHAPTER 15 Session Procedures, Mastering, and CD Burning
PQ subcode editing. Some programs that do this are Sony Creative
Software CD Architect, Adaptec Jam, and Goldenhawk Technology
CDRWIN. Some DAW software includes a CD-burning application
that lets you set start IDs.
Here’s a suggested procedure to master an album:
1. If you already recorded your mixes directly onto hard disk as
stereo WAVE files, skip to Step 4.
2. If your song mixes are WAVE files on a CD-R, put the CD-R in
your CD-ROM drive and copy each song to your hard drive.
Then go to Step 4. If your song mixes are audio tracks (.cda
files) on a CD-R, convert each track to a WAVE file on your
hard drive using ripper software. Skip to Step 4.
3. If your mixes are on a stand-alone 2-track recorder without a
USB or FireWire port, plug the recorder’s digital output into
the digital input of your audio interface. If your interface lacks
a digital input, use analog out to analog in. Launch the recording software. Set levels if necessary, start recording on your
hard drive, and play the 2-track recording containing your
mixes. All the mixes will copy in real time to your hard drive
to create a single long WAVE file.
If your 2-track recorder has a USB or FireWire port, plug it into
your computer’s matching port. The recorder will appear as
a hard drive to your computer. Copy the WAVE file(s) of the
mixes to your computer hard drive.
4. Now that all the song mixes are on your hard drive, launch your
DAW software. Import the first song’s WAVE file to a track at time
zero. The waveform of the song appears as a clip or segment of
audio. Slip-edit or trim the beginning and end of the song clip to
remove extra space and noises. Repeat for the other songs.
Click on the trimmed Song 1 and slide it 10 frames to the right in
your mastering software. That way the CD player won’t miss the
beginning of Song 1.
It’s convenient to put each song clip on a different track, one after
another (Figure 15.4). That way you can easily adjust the spacing
between songs, and apply different fader settings and processing to
each song as needed. Assign all the tracks to the same stereo bus so
you can apply bus processing to them later.
Session Procedures, Mastering, and CD Burning CHAPTER 15
FIGURE 15.4
Placing song clips on successive tracks makes mastering easier.
5. Now that all the songs are in place and trimmed, add a fadeout at the end of certain songs if desired. If your CD-burning
program can write PQ subcodes (start IDs), you can crossfade between two songs: overlap their clips near the transition
point, and use the crossfade function in your DAW. Or fade
out the end of one song and fade in the beginning of the next
(as shown in Figure 13.17).
6. If your CD-burning program can write PQ subcodes (start
IDs), you can adjust the spacing or gap between songs. Two
to three seconds of silence between songs is typical, but go by
ear. Use a longer space if you want to change moods between
songs. Use a shorter space to make similar songs flow together.
A short space also works well after a long fade-out because the
fade-out acts like a long pause between songs.
7. Click on and play part of each song’s waveform to check for
loudness. Make all the songs equally loud by adjusting the
track’s fader for each song. If you put all the songs on the same
stereo track, set up an automation volume envelope.
373
374
CHAPTER 15 Session Procedures, Mastering, and CD Burning
Here’s one method to match song levels. If most songs are
equally loud, set the fader to zero before each of those songs
in order to avoid processing. Then turn down any louder songs
and turn up quieter songs to match the rest of the songs. Do
this by ear. You might use post-fader rms metering on the stereo mix bus as a rough indicator of average levels.
Caution: If you increase a song’s level, make sure the peaks in
the song don’t clip.
Another way to match song levels is to find the song with the
highest peaks, and leave its level alone Adjust the levels of
other songs so they match the loudness of the song with the
highest peaks. Again, do this by ear. You may need to turn up
the intro of a song so it works in context with the previous
song. Consider using Harmonic Balancer (www.har-bal.com),
a tool to match song levels and tonal balances.
8. Apply EQ to songs that need it. Put songs with different EQs on
different tracks. If necessary, touch up levels after adding EQ.
9. If desired, apply multiband compression, limiting, and normalization to the mix bus in order to get a “hot” or loud CD.
Don’t overdo it, or your project will be distorted and fatiguing
to listen to. Squashed dynamics can suck the life out of music.
If you apply only peak limiting and normalization, you’ll get a
hot CD without altering the musical dynamics. Play audio clip
40 at www.elsevierdirect.com/companions/9780240811444.
The idea is to knock down the peaks in the waveform because they
don’t contribute to perceived loudness—the average level does. Once
you limit the peaks about 6 dB, you can normalize (raise the overall
level) and thus create a louder program. Normalizing to 100% of full
scale can create errors with some D/A converters, so you might normalize to 96% or ⫺0.3 dB.
THE LOUDNESS WARS
Musicians and record labels, more than recording engineers or record
producers, are requesting hotter levels on CDs. One reason is that a lowlevel CD played with others in a CD changer tends to sound less exciting.
Musicians also want their songs to stand out on the radio.
Session Procedures, Mastering, and CD Burning CHAPTER 15
Loudness depends mostly on average (rms) level rather than peak level. So
in an attempt to raise average levels, engineers limit or clip the peaks. Overall
compression of the mix is common too. Unfortunately, the song loses its
dynamic range and its musical excitement.
When I master a pop-music CD, I often compromise by limiting peaks by no
more than 7 dB, then normalizing without overall compression or clipping. That
retains the dynamic range but loses a little transient clarity.
If playback equipment had switchable automatic volume control
(compression), then we wouldn’t need to maximize levels in the recording.
According to mastering engineer Bob Katz, hot CDs and quiet CDs end up
at the same level when processed with radio-station compressor/limiters.
Over-compressed CDs don’t sound louder on the air; they just sound more
distorted. So ignore the “loudness wars” and avoid excessive compression.
Bob recommends that pop music be mastered at an average or rms level of
⫺14 dBFS while listening at 83 dBSPL C-weighted, slow meter reading. He
suggests an average level of ⫺20 dBFS for classical music.
10. After your musical program is edited, you have two options:
(1) If you have professional CD-burning software that can
write PQ subcodes, export (save) the mastered song mixes
to a single 16-bit/44.1 kHz stereo WAVE file. If your recording was 24-bit, first turn on dithering. That retains much of
the 24-bit sound quality when converting to 16-bit format.
(2) If you have consumer software that burns a CD from a list
of song WAVE files, click on one song’s clip in your DAW to
highlight it, then export the mix to a single 16-bit/44.1 kHz
stereo WAVE file. If your recording was 24-bit, first turn on
dithering. Repeat for all the song clips.
TRANSFERRING THE MASTERED PROGRAM
TO CD-R
At this point your hard drive contains either one long WAVE file of
all the mastered songs, or several WAVE files, each one a mastered
song. If the latter, launch your CD-burning application and drag the
375
376
CHAPTER 15 Session Procedures, Mastering, and CD Burning
WAVE files in order to the playlist. To prevent errors on the CD, burn
it at 4 ⫻ speed and don’t multitask. Check the finished CD in a CD
player.
Suppose you have pro CD-burning software. It creates CD tracks
from a single WAVE file by following a cue sheet: a text file that lists
the start time of each song. When the CD is recorded, the software
creates a Start ID for each song based on the cue sheet. With this
method you can create CD song-start IDs that occur even during a
continuous program, such as a live concert recording.
CDRWIN at www.goldenhawk.com is such a program. I’ll describe
how to use it as an example.
1. Note the start time of each song in the DAW editing software’s
Edit Decision List or playlist. Or check the start time of each
song clip by right-clicking on it or by looking at its beginning
on the project time line. The start time of each song is specified in your DAW in minutes:seconds:frames, and there are
30 frames per second. Write down the start time of each song.
2. In the cue sheet, make each song’s start time about one-third
of a second (10 frames) before the actual start time of each
song. That way the CD player’s laser will have time to find the
track before playing audio. For example, if a song actually starts
at 12:47:28, make its start time in your cue sheet 12:47:18.
List Song 1’s start time as 00:00:00. The cue sheet should
also include the name and filepath of the mastered program’s
WAVE file.
3. Place a blank CD-R in your CD burner. Don’t put a paper
label on the CD before recording because the label can cause
jitter.
4. Open your CD-burning program and load in the cue sheet
text file. Set the recording speed to 2 ⫻ to 4 ⫻ speed to reduce
errors. Some CD-recorders and blank CD-Rs have fewer errors
at higher speeds.
5. Start recording the CD. The CD-burning program writes the
cue sheet’s start times in the CD-R’s table of contents, and copies the mastered program’s WAVE file to the CD-R. Each CD
track starts at the time you specified in your cue sheet. To prevent glitches, don’t multitask while the CD is burning.
Session Procedures, Mastering, and CD Burning CHAPTER 15
6. When the CD is finished, handle it by its edges to prevent
error-causing fingerprints. Play the CD from start to finish to
check for glitches. Press the track-advance button to make sure
each song starts at the right time. If you look at the CD’s contents in your computer, you’ll see a separate .cda (CD-Audio)
track for each song.
You can identify the CD by writing on the label side with a felt-tip
marker. To avoid damaging the CD plastic layer, use water-based
Sharpie or Memorex CD marking pens, available in the CD-R section
of many stores. Some software can print liner notes, including the
track numbers, titles, and timing. If you’re duplicating CDs yourself,
you might want to use a CD printer rather than labels. The adhesive
in labels can damage the top CD layer or warp the CD over time,
making the CD unplayable.
Another way to create master CDs is to use the Alesis MasterLink: a
CD recorder with a built-in hard drive and mastering tools. With it
you can record your stereo mixes, edit and master them, and burn
a finished CD. Sample rates can be up to 96 kHz and word length
can be 16-, 20-, or 24-bit on Masterlink’s CD24 format discs. The
MasterLink includes sample-rate conversion and noise shaping to
alter sample rates and bit resolution as needed.
Several editing/mastering tools are offered with MasterLink, such as
16 different playlists, gain control, editing start and end points, joining or splitting song sections, EQ, compression, normalization, and
peak limiting.
CD-Text and ISRC Code
You can add non-audio data to your CD when you create the cue
sheet. If you want the listener to see on-screen each song’s title and
performer as the CD plays, create a CD-Text file using the CD-Text
editor that comes with most CD-burning software. Include the filepath of the CD-Text file in the cuesheet. Not all CD players support
CD Text.
Some players get CD-Text information from an online database
rather than from the CD. You upload the CD-Text information to the
377
378
CHAPTER 15 Session Procedures, Mastering, and CD Burning
database so that the player can display it. For example, iTunes needs
to access an online CD database called Gracenote (formerly CDDB)
to get the artist and song titles. Gracenote is at www.gracenote.com.
Every database has a different procedure for uploading CD-Text.
When using the iTunes database,
1. Insert your CD and open iTunes.
2. Select all tracks in the iTunes window.
3. Select Get Info and enter the album name, artist, year, genre,
copyright info, and so on.
4. Select each track individually, select Get Info, and enter the
track’s correct name.
5. Go to Advanced ⬎ Submit CD track names. Your information
will upload to the Gracenote database. When someone puts
your CD into iTunes, your info will show up.
When using the Windows Media Player database, send a copy
of your CD to AMG (All Media Guide) at www.allmediaguide
.com/labels.html. Many players access the freedb database at www
.freedb.org. Go there to see a list of supported players you can use
to submit CD information. Popular applications are Exact Audio
Copy and Nero. Another database is Muse at www.muze.com/html
/industry.
iTunes before version 7 didn’t display CD-Text. iTunes 7 can display CD-Text information, but only if it has been uploaded to the
Gracenote database.
The International Standard Recording Code (ISRC) is a 12-character
code that identifies each track. It’s a tool for royalty collection
and anti-piracy. Information on applying for ISRC codes is at www
.recordlabelresource.com/isrc-registration.html and www.riaa.com.
If the CD tracks have ISRC codes, include them in the cue sheet as
well. Some online music providers, such as CDBaby, can create the
ISRC codes for you.
MASTER LOG
Type or print a log describing the CD-R master (such as shown in
Table 15.2) and include it with your master.
Session Procedures, Mastering, and CD Burning CHAPTER 15
Table 15.2
A CD Master Log
CD Master Log
Album title: Don’t Press That Button (demo)
Artist: Puff Daddy Bartlett
Mastering date: 2-26-06
Track
Start MM:SS:FF
Duration
Title
1
00:00:00
03:23
60 Gigabytes to Go
2
03:24:20
03:15
Unplugged Plug-In
3
06:41:00
02:49
Digital Dropout
Blues
4
09:32:05
02:55
Late and See
5
12:29:11
03:07
Buffer the Vampire
Total running time: 15:47
Also include the CD liner notes such as song lyrics, instrumentation,
composers, credits, arrangers, publishers, copyright dates, and artwork. Include contact information for the artist, engineer, and mastering engineer. Stay in contact with the duplication or replication
house, especially about artwork.
Before you send out the master, be sure to make another copy that
you keep in your studio. This copy can be used if the master is lost
or damaged.
Note that the CD-R pre-master doesn’t leave the studio until all studio time is paid for! When this is done, send the CD-R to the duplication house.
The document “Recommendations for Delivery of Recorded Music
Projects” provides examples of session sheets, track sheets, and so
on. It also recommends media for delivering and backing up master recordings. It can be found at www.aes.org/technical/documents.
Search for AESTD1002.1.03–10_1.pdf.
It’s amazing how the long hours of work with lots of complex equipment have been concentrated into that little CD-R—but it’s been
fun. You crafted a product you can be proud of. When played, it
recreates a musical experience in the ears and mind of the listener—
no small achievement.
379
380
CHAPTER 15 Session Procedures, Mastering, and CD Burning
COPYRIGHTS AND ROYALTIES
It’s important to copyright original songs so that others can’t claim
them as their own. Go to www.copyright.gov/forms to download
form SR (for Sound Recordings) and form CON (continuation of
form SR to add more songs). The filing fee is $45 for submission
by regular mail or $35 by electronic submission. That fee lets you
copyright several songs by the same composer(s) as a single work on
one CD. Just be sure to list all the song titles. Send the forms, CD,
and fee in a box to the U.S. Copyright Office, address at the above
Web site.
If you record and sell cover tunes (songs that have been recorded
by other artists), or if you sample other artists’ music in your own
recording, you must obtain a “mechanical license” and pay a royalty
fee to BMI or ASCAP. The Harry Fox Agency (HFA) is the premier
musical industry resource for licensing. They offer a service called
Songfile, a simple, fast way to obtain licenses for 25 to 2,500 copies
of a recording as a CD, cassette, LP, or digital download. If the cover
song is under 5 minutes long, the royalty fee is about $45 per song
for every 500 CDs that you distribute. For details check out www
.harryfox.com, www.bmi.com, and www.ascap.com.
CHAPTER 16
The MIDI Studio:
Equipment and Recording
Procedures
Welcome to the computerized world of MIDI equipment—samplers,
synthesizers, sampling keyboards, drum machines, and sequencers.
Because other texts explain this equipment in detail, brief definitions
serve the purpose here. See Appendix D for books on MIDI.
Musical Instrument Digital Interface (MIDI) is a standard connection
between electronic musical instruments and computers that allows
them to communicate with each other.
Here are some things that MIDI lets you do:
■
■
■
■
■
■
Play a piano-style keyboard, drum pads, MIDI guitar, or breath
controller to produce a sound like any instrument.
Record and play back a musical performance that is independent of the sound of that performance. You can hear your performance played by any musical instrument or synthesized
sound, or change the instrument sound after your performance
is recorded.
Record a difficult part slowly, or one note at a time, and play it
back at any tempo you wish.
Enter the notes on a musical scale or on a piano-roll grid, one
at a time, and play them back at any tempo.
Edit any note of the performance—change its timing, length, or
pitch.
Create the effect of a band playing. Record a performance that
is played by drum samples, overdub another performance that
is played by a synthesized bass, overdub another performance
played by piano samples, and so on.
381
382
CHAPTER 16 The MIDI Studio: Equipment and Recording Procedures
■
■
Combine or layer the sounds of two electronic musical instruments by playing them both with the same performance.
Quantize the notes to make them rhythmically precise.
The MIDI signal is a stream of digital data—not an audio signal—
running at 31,250 bits per second. It sends information about the
notes you play on a MIDI controller, such as a piano-style keyboard
or drum pads. Up to 16 channels of information can be sent on a
single MIDI cable.
There are three types of MIDI ports on MIDI devices:
1. MIDI IN receives data going into the device.
2. MIDI OUT sends out data generated by the device.
3. MIDI THRU is like MIDI OUT, but duplicates the data that are
at the MIDI IN port.
Connect MIDI OUT from the sending device to MIDI IN of the
receiving device. For example, connect a keyboard controller’s
MIDI OUT to the MIDI IN connector of a MIDI/audio interface, a
sequencer, or a sound module. Use MIDI THRU to connect two or
more receiving devices in a row. For example, connect a keyboard
controller’s MIDI OUT to sequencer MIDI IN, and connect sequencer
MIDI THRU to sound module MIDI IN.
See MIDI Clock and MIDI Timecode in the Glossary.
MIDI-STUDIO COMPONENTS
The following equipment typically is used in a MIDI studio:
■
■
■
■
■
■
■
■
■
■
■
■
MIDI controller
Sequencer
Synthesizer
Sampler and sample CDs
Drum machine or soft synth playing drum-set samples
Power amplifier and speakers, or powered speakers
Personal computer
MIDI computer interface, or audio interface with MIDI I/O
Mixer (optional)
Recorder–mixer (optional)
Effects or plug-ins
Audio cables
The MIDI Studio: Equipment and Recording Procedures CHAPTER 16
FIGURE 16.1
M-Audio Oxygen 8v2, and example of a MIDI keyboard controller.
■
■
MIDI cables
Equipment stand
You have learned about most of these in previous chapters, but a
review might help at this point.
A MIDI controller is an instrument that generates MIDI data when
you play on it. Examples are a piano-style keyboard (Figure 16.1),
drum pads, MIDI guitar, or a MIDI breath controller. A synthesizer
or drum machine can act as a controller.
Some keyboard controllers are compact and lightweight, and connect to your computer via a USB cable. Examples are Korg’s 49-key
Kontrol49, M-Audio’s 25-key Oxygen 8v2, E-MU’s Xboard Series,
Edirol’s 49-key PCR-500, and Novation’s ReMoteslCo25. Because
they have fewer keys than a standard keyboard, they use octave up/
down buttons.
A sequencer is a device or program that lets you record, edit, and play
back MIDI data. A recording done on a sequencer is called a sequence,
which is a MIDI song. Unlike an audio recorder, a sequencer doesn’t
383
384
CHAPTER 16 The MIDI Studio: Equipment and Recording Procedures
FIGURE 16.2
A screen shot of a piano-roll view, part of a MIDI sequencer program.
record audio. Instead, it records the key number of each note you
play, note-on signals, note-off signals, and other parameters such as
velocity, pitch-bend, and so on. A sequencer captures a performance,
not the sound. Then any sound you want can be played by that performance. You can edit the recorded performance (sequence) to fix
wrong notes, etc.
A sequencer records and plays MIDI files (.mid files). They are
sequencer recordings of your own performances, or are MIDI files
that can be downloaded from the Web.
The sequencer can be a stand-alone unit, a circuit built into a keyboard instrument, or a computer running a sequencer program
(Figure 16.2). Like a multitrack audio recorder, a sequencer can
record 8 or more tracks, with each track containing a performance of
a different instrument.
A synthesizer is a musical instrument that creates sounds electronically. It can play MIDI data, either from a built-in keyboard, a separate MIDI controller, a MIDI sequence, or a MIDI file downloaded
from the Web. Synthesizers come in four forms: piano-style, sound
module, software, or a synth chip on a sound card.
■
■
A piano-style synth has a piano keyboard and built-in sound
generators. You might want to use more than one synth to
expand your palette of sounds.
A sound module or tone module is a synthesizer without a keyboard. This stand-alone device is triggered by a sequencer or a
MIDI controller.
The MIDI Studio: Equipment and Recording Procedures CHAPTER 16
■
■
A soft synth or virtual instrument is a synthesizer that is simulated in software. It runs in your CPU. The GUI of the software
looks like a hardware synthesizer. A wavetable soft synth plays
samples of real instruments, which sound more realistic than
an FM soft synth.
Another option is a synth chip, which is built into many sound
cards.
Some examples of soft synths are E-MU Proteus, Arturia ARP,
Cakewalk Dimension Pro and Pentagon, McDSP Synthesizer One,
Native Instruments Reaktor, and TASCAM GigaStudio Orchestra.
The musical sounds that synthesizers make—such as a fretless bass,
grand piano, sax, or drum set—are called patches, programs, or
presets. A multitimbral synthesizer can play two or more patches
at once. A polyphonic synthesizer can play several notes at once
(chords) with a single patch.
A sampler is a device that records sound events, or samples, into computer memory and plays them back when activated by a sequencer
MIDI file or MIDI controller. A sample is a digital recording of one
note of a real sound source: a flute note, a bass pluck, a drum hit,
etc. A sample also can be a digital recording of a short segment of
another recording. The sampling process is described in Chapter 9
under Digital Recording.
A soft sampler is software that plays samples and lets you map
them along your keyboard. In other words, you can tell the sampler
which notes on the keyboard play which samples. A sampler lets
you import any sound, such as a WAVE file of a single note that you
recorded, or a file from a sample CD. In contrast, a soft synth generates its own sounds—either the ones supplied with it, or your own
customized versions.
Some software sample players are Korg Legacy Collection, IK
Sampletank, Native Instruments Kontakt 2, and Steinberg HALion.
Also check out these Web sites: www.synthzone.com/softsyn.htm,
www.sonicspot.com/softwaresynth.html, www.hitsquad.com/smm/,
and www.kvraudio.com.
A soundfont is an audio sample (an instrument patch) in a special
.sf2 format. It’s like a WAVE file, but also includes a key range so
that when you play a MIDI note number (keyboard key), it plays the
385
386
CHAPTER 16 The MIDI Studio: Equipment and Recording Procedures
sample pitch assigned to that note number. Soundfonts also include
velocity switching, note envelope, looping, release sample, filter, and
low-frequency oscillator (LFO) settings. A single soundfont can contain many WAVE files of different pitches. You can import a soundfont into a sample player. To use soundfonts, you need an E-MU
or SoundBlaster sound card, a MIDI controller, MIDI interface, and
MIDI sequencing software.
Often a sampler is built into a sample-playing keyboard, which
resembles an electronic piano. It contains samples of several different musical instruments. When you play on the keyboard, you hear
the sample notes. The higher the key you press, the higher the pitch
of the reproduced sample.
You can buy CDs with samples for use in your own projects. You
copy the samples to your hard drive, load them into a software sample player, then trigger them with a sequencer or MIDI controller. For
example, TASCAM’s GigaStudio is sample-playing software with a
huge sample library. FXpansion’s BFD is a drum sample library. Other
libraries are available from www.MI7libraries.com and www.sonomic
.com. Several DAW recording programs have samples included.
You can also download samples from the Web. Samples are included
in Steinberg’s HALion Player (www.steinberg.net) and Maxim Digital
Audio Piano (freeware at http://mda.smartelectronix.com).
A drum machine or beat box is a device that plays built-in samples
of all the sounds of a drum set and percussion (Figure 16.3). It also
is a sequencer that records and plays back drum patterns played or
programmed with built-in keys or drum pads. Some units can sample sounds. Most recorder–mixers have a drum machine built in. In
a DAW, you might use a soft sampler that plays drum samples—a
virtual drummer.
Some drum plug-ins for MIDI/audio recording programs include
EastWest Storm Drum, Toontrack Drumkit, Steinberg Groove Agent,
M-Audio Drum & Bass Rig, Submersible DrumCore, and Cakewalk
Session Drummer.
A power amplifier and speakers (or powered speakers) let you hear
what you’re performing and recording. Usually these are small monitor speakers set up in a Nearfield arrangement (about 3 feet apart
and 3 feet from you).
The MIDI Studio: Equipment and Recording Procedures CHAPTER 16
FIGURE 16.3
BOSS DR-880, an example of a beat box.
A computer is used mainly to run a sequencer program, which
replaces the stand-alone sequencer and its tiny LCD screen. The computer monitor screen displays much more information at a glance,
making editing easier and more intuitive.
A step up from a sequencer program is a MIDI/audio recording program. It records both MIDI sequences and digital audio tracks on
your hard drive, and keeps them in sync. In other words, this program
lets you add microphone signals such as a vocal, sax, etc., to MIDI
sequences. This software and your computer form a Digital Audio
Workstation (DAW). Details are given in Chapter 13, Computer
Recording.
An alternative to a computer with MIDI software is a recorder–mixer
(hardware DAW). It combines a multitrack audio recorder, MIDI
sequencer, and mixer into a single chassis.
387
388
CHAPTER 16 The MIDI Studio: Equipment and Recording Procedures
A librarian program organizes your collection of samples and synth
patches.
A voice editor program lets you create your own synth patches.
A notation program converts your performance to standard musical
notation. You can edit the notes, add lyrics and chords, and print out
a copy.
A MIDI interface plugs into a user port in your computer and converts MIDI signals into computer data so that it can be recorded,
edited, and played back with sequencer software. Many sound cards
and audio interfaces include a MIDI interface.
If you have two or more hardware synthesizers, or a synth and a
drum machine, you need a line mixer to blend their audio outputs
into a single stereo signal.
Audio effects include compression, reverb, echo, gating, chorus, and
so on. MIDI effects are effects that process MIDI signals rather than
audio signals. Some MIDI effects are arpeggiate, transpose, and delay.
Audio cables carry audio signals and typically have a 1/4-inch phone
plug on each end. They connect synths, sound modules, and drum
machines to the line inputs in your mixer or audio interface.
MIDI cables carry MIDI signals and are used to connect synths, drum
machines, and computers together so that they can communicate
with each other. A MIDI cable is a 2-conductor shielded cable with a
5-pin DIN plug on each end. Pins 4 and 5 are the MIDI signal, pin 2
is shield, and pins 1 and 3 aren’t connected.
An equipment stand is a system of tubes, rods, and platforms that supports all your equipment in a convenient arrangement. It allows comfort, shorter cable lengths, and more floor area for other activities.
A keyboard workstation (Figure 16.4) includes several MIDI components in one chassis: a keyboard, a sample player, a sequencer, and
perhaps a synthesizer, and disk drive. That’s everything you need to
compose, perform, and record instrumental music. Many workstations include drum sounds so that you can get by without a separate
drum machine. A few include a mic input so you can record vocals
or acoustic instruments.
The MIDI Studio: Equipment and Recording Procedures CHAPTER 16
FIGURE 16.4
The Korg M3, an example of a keyboard workstation.
The rest of this chapter describes recording procedures for some
MIDI studio setups. You’ll also need to read your instruction manuals thoroughly and simplify them into step-by-step procedures for
various operations. Note that each piece of MIDI gear has its own
idiosyncrasies, and the instructions may have errors or omissions. If
you have questions, call or e-mail tech support for your equipment.
Also look for books and videos devoted to your specific musical
instrument.
RECORDING MUSIC MADE BY SOFT SYNTHS
In this method, you will play a MIDI controller and record the performance on a MIDI track in a MIDI/audio recording program on
your computer. Software synths, samples, and drums in your computer will play the sounds during recording and playback (Figure
16.5). They are included with most recording software, or you can
buy them separately.
389
390
CHAPTER 16 The MIDI Studio: Equipment and Recording Procedures
FIGURE 16.5
rgc:audio Pentagon, an example of a soft synth.
The MIDI Signal Chain
Figure 16.6 shows a MIDI signal chain that uses a soft synth. From
start to finish (left to right), these are the components and their
functions:
1. MIDI controller. A piano-style keyboard, drum pads, or
breath controller that outputs MIDI code (note on, note off,
key number, etc.) when you play it. The code is tagged with a
channel number from 1 to 16, which you set in the controller.
2. A MIDI cable from your controller’s MIDI OUT to your interface’s MIDI IN.
3. A MIDI interface or MIDI/audio interface. It converts the
MIDI signal into a data format that your computer can
understand.
4. A PCI, USB, or FireWire connection from the interface to the
computer.
The MIDI Studio: Equipment and Recording Procedures CHAPTER 16
MIDI
controller
PCI, USB,
MIDI/audio FireWire
interface
Sequencer Soft
software synth
Record
MIDI
out/in
Computer
PCI, USB,
FireWire
Play
MIDI
sequence
Audio
Monitors
MIDI/audio
interface
FIGURE 16.6
A MIDI signal chain that uses a soft synth.
5. A MIDI driver (not shown). A small program in your computer that allows your recording software to transfer data to
and from an audio/MIDI interface.
6. Sequencer (MIDI/audio) software running in your computer.
It records the MIDI signal to the computer’s hard drive and
plays it back.
7. Soft synth. A plug-in in your MIDI/audio host program. As
you record your performance, you’re listening to the software
synthesizer play music according to your keyboard presses. So
in addition to being recorded, the MIDI signal goes to a soft
synth or sample player of your choice.
8. PCI, USB, or FireWire connection (same as #4 above). The
synth’s digital audio is sent back to your MIDI/audio interface
through the same PCI, USB, or FireWire cable that is sending
MIDI data to the computer sequencer.
9. The interface (same interface as #3 above) converts the
synth’s digital audio to analog, and sends that musical signal
to its stereo output channels.
10. Monitors. The stereo output of your interface feeds amplified
speakers or headphones so you can hear the sound of your
performance.
During playback, the recorded MIDI sequence “plays” or triggers the
soft-synth patch, and its audio is routed through your interface and
monitors.
391
392
CHAPTER 16 The MIDI Studio: Equipment and Recording Procedures
MIDI Recording Procedure
Now that you understand the MIDI signal chain, the steps below will
make more sense. Here’s how to create the audio and MIDI tracks in
your sequencer (MIDI/audio) program.
1. Set Up the Audio and MIDI Tracks
Insert a MIDI track to record your MIDI sequence, and insert
an audio track that contains the soft synth as a plug-in. (Your
sequencer might combine the MIDI and audio tracks into one
track).
Synth and sample patches are stored in groups called banks, and
each bank contains several different patches In the MIDI track,
choose a synth bank and patch (bass, drums, piano, etc).
In the MIDI track input, specify the MIDI channel that the synth
will respond to: either Omni mode (any channel) or the same
channel number that your MIDI controller is set to.
In the MIDI track output, specify the synth or sample player that
you want the MIDI track to play.
Set the audio track output to the stereo output of your interface.
2. Start Recording
In your interface control program, or in your recording program,
set the latency or buffer size low (under 4 msec if possible) while
tracking so that the synth responds without a noticeable delay.
See Appendix B under the heading Minimizing Latency.
Choose a tempo in your sequencer. Set the metronome to count
off two measures.
Enable record mode on the MIDI track. Using your mouse, click on
the Record button on-screen or on your computer keyboard. You’ll
hear the sequencer’s metronome ticking at the tempo you set.
3. Play Music on Your Keyboard
Listen to the metronome and play along with its beat. The
sequencer keeps track of the measures, beats, and pulses. As you
play, MIDI data from your keyboard goes from keyboard MIDI
OUT to interface MIDI IN, and is recorded as a MIDI file or
sequence. When the song is done, click on STOP.
Another way to record your performance is in step-time, one note
at a time. You also can set the sequencer tempo very slow, record
while playing the synth at that tempo, and then play back the
sequence at a faster tempo.
4. Play Back the Sequencer Recording
The MIDI Studio: Equipment and Recording Procedures CHAPTER 16
Click on PLAY. You’ll hear the sequence playing your soft synth. If
you change the patch and click on PLAY, you’ll hear the same performance played by a different instrument.
5. Layer the Drum Track
Using the recording software’s sound-on-sound setting, you can
record several passes of a MIDI drum track, adding another instrument each time. On the first pass tap-in the kick and snare, on the
next pass add hi-hat, and so on.
6. Quantize the Track
Quantizing automatically corrects the timing of each note
to the nearest note value (quarter note, eighth note, and so on).
If you wish, quantize the performance by the desired amount.
Caution: Quantizing can de-humanize the performance, making it too rhythmically perfect. It’s better to adjust the timing of certain notes only, and to the smallest note value that
works. Quantizing is essential if you want to use a notation
program.
7. Punch-In/Out to Correct Mistakes
To correct a musical error, you can punch-in to record mode
before the mistake, record a new performance, and then punchout of record mode. Another way to correct errors is to edit the
MIDI notes in the piano-roll view, described in Step 8, Edit the
Sequence Recording. Here’s one way to punch in:
1. Go to a point in the song a few bars before the mistake.
2. Just before you get to the mistake, punch-in to record mode
and play a new, correct performance.
3. As soon as you finish the correction, punch-out of record
mode.
Alternatively, you can use autopunch. With this feature, the computer punches in and out automatically at preset measures; all you
have to do is play the corrected musical part. Perform an autopunch
as follows:
1. Using the computer keyboard or mouse, set the punch-out
point (the measure, beat, and pulse where you want to go out
of record mode).
2. Set the punch-in point (just before the part you want to correct).
3. Set the cue or pre-roll point (where you want the track to start
playing before the punch).
393
394
CHAPTER 16 The MIDI Studio: Equipment and Recording Procedures
4. Click on PLAY.
5. When the screen indicates punch-in mode, or when the appropriate measure comes up, play the corrected part.
6. The sequencer punches out automatically at the specified point
in the song.
These punch-in routines were done in real time. You can also punchin/out in step-time:
1. Go to a point in the song just before the mistake.
2. Set the sequencer to step-time mode.
3. Step through the sequence note by note, and punch-in to
record mode at the proper point.
4. Record the proper note in step-time.
5. Punch-out of record mode.
8. Edit the Sequence Recording
You might find it easier to edit the MIDI performance. Go to the
MIDI edit screen or piano-roll view (Figure 16.2). It’s a grid showing pitch verses time. The pitch of each note is represented by its
height on the grid, and the duration of each note is represented by
it length. You can grab incorrectly played notes and put them at the
correct pitch and timing, delete unwanted notes, copy and paste
phrases of notes, stretch or shorten note durations, and so on.
When you’re finished editing one MIDI performance, overdub
other MIDI tracks, edit them, and set up a mix.
9. Arrange the Song by Combining Sequences
At this point you can put together your composition. Many
songs have repeated sections: The verse and chorus are each
repeated several times. If you wish, you can record the verse and
chorus once. Then copy the verse section and paste it every place
it occurs in the song. Do the same for the chorus.
10. Add Audio Tracks
Suppose you want to add vocals or live instruments to the mix.
Here’s a suggested procedure.
1. If you’re using a sound card, plug a mic into a mic preamp or
mic input of a mixer. Connect the preamp or mixer line output to your sound-card line input. If you’re using an audio
interface, plug a mic into one of its mic connectors. Connect
the interface’s PCI, USB, or FireWire port to your computer.
The MIDI Studio: Equipment and Recording Procedures CHAPTER 16
2. In your recording software, insert an audio track, and set
its input source to the audio interface or sound-card channel that the mic signal is plugged into. For example if you
plugged a mic into interface input 3, set the track’s input to
interface input 3. Enable Record mode on that track.
3. Set the recording level by adjusting the microphone input
trim or gain on your preamp, mixer, or interface.
4. Go to the beginning of the tune and hit PLAY in your DAW.
The MIDI sequences that you recorded earlier should start
playing. (You may need to press the PLAY key on a drum
machine first if it is an outboard device, rather than a part of
software.)
5. While listening to the MIDI tracks playing through headphones, record the vocal on an audio track. Then overdub
more vocals and non-MIDI instruments on other open tracks.
You might record a few takes and then cut and paste selected portions to create a perfect take. For example, record one good chorus
and copy it in each chorus section in your song. You also can edit
audio notes to correct their pitch or timing.
When you play back what you recorded, the recorded MIDI tracks
play the soft synths, and the digital audio tracks play their audio signals. A mix of the soft synths and audio tracks plays from the audio
interface line output, which is connected to your monitors.
11. Mix, Export the Mix, and Burn a CD
1. After all your tracks are recorded, use the on-screen mixer to
set up a mix of the audio tracks and MIDI instruments. Adjust
levels, panning, and effects (plug-ins).
2. Play the song several times to perfect the mix and to set up
automation. See Chapter 12 for mixing procedures and
automation.
3. Export the mix to a stereo WAVE file on your hard drive. Then
you can burn a CD of the song or convert it to an MP3 file.
RECORDING A HARDWARE SYNTH
Suppose you like the sound of a hardware synth. You want to record
its sound and also record a MIDI sequence of your performance so
you can edit it. Figure 16.7 shows the connections.
395
396
CHAPTER 16 The MIDI Studio: Equipment and Recording Procedures
COMPUTER
SYNTHESIZER
MIDI / AUDIO
RECORDING SOFTWARE
PCI, USB or
FIREWIRE
MIDI IN
MIDI OUT
MIDI OUT
MIDI IN
LINE OUT
AUDIO MIX OUT
TO MONITORS
LINE IN
AUDIO/MIDI
INTERFACE
FIGURE 16.7
Setup for recording a hardware synth on a computer.
1. Monitor a mix of the synth audio and sequencer playback. In
other words, monitor a mix of the interface input signal and
output signal.
2. Insert a MIDI track in your sequencer program. Set its input
to your MIDI/audio interface in Omni channel mode. Set the
output of the MIDI track to your MIDI/audio interface. That
way, during playback the recorded sequence will pass through
the MIDI/audio interface to your hardware synth and will trigger it to play music.
3. Enable Record mode in the MIDI track. Record your performance on the MIDI track. (The sequencer will also record
whatever MIDI channel number is set in the synth.)
4. If necessary, set your synthesizer to External Clock (MIDI
clock) so it will follow the timing of the recorded sequence.
5. Play the sequence and edit it in the piano-roll view.
6. Insert an audio track. Set its input to the interface line input
that is connected to the synth audio output. Disable Record
mode on the MIDI track, and enable record mode on the
audio track.
7. Make sure that the synth MIDI channel is unchanged. Set levels and start recording. In other words, play the MIDI track
The MIDI Studio: Equipment and Recording Procedures CHAPTER 16
while recording the synth audio on the audio track. The MIDI
sequence will “play” or trigger the synthesizer as you record its
output on an audio track.
8. When the song is done, you have an audio track produced
by the hardware synth. If you wish, you can disable the MIDI
track that drove the synth.
“NO SOUND” MIDI TROUBLESHOOTING
Suppose you load a MIDI file and hit PLAY on your sequencer, but
you hear nothing. Or you play notes on your MIDI controller, but
there’s no sound. Not a peep.
That’s not too surprising, considering that the MIDI signal chain is a
long one. Every link in the chain has to work, whether it’s hardware,
software, or their settings. I’ll give some MIDI troubleshooting advice
that should coax your system to speak when you ask it to.
If any part of the chain isn’t turned on, or has the wrong settings,
you won’t hear a sound when you monitor the synth—no matter
how hard you pound the keys! Listed below are some possible reasons why your synth can’t sing, and what to do about it.
■
■
■
■
MIDI OUT or MIDI THRU isn’t connected to MIDI IN somewhere in your system. Trace the cable connections from beginning to end, and see whether a MIDI cable isn’t plugged in
where it should be. The cable itself might be broken.
Your MIDI interface isn’t communicating with your computer.
A missing or flaky PCI, USB, or FireWire connection can disrupt the data flow between interface and computer. Re-plug or
replace the cable. You might need to restart your sequencer program too. Some recording software has an on-screen indicator
that flashes when the software is receiving MIDI data, so check
it out.
When you’re tracking, the MIDI sequencer track isn’t selected.
Some sequencer software doesn’t monitor the soft synth unless
you select its MIDI track. In some sequencers, you can livemonitor only one synth at a time, but you can play back multiple synth tracks at once.
The MIDI sequencer track has no MIDI channel assigned, or
is set to a different channel than the MIDI controller. Just as
397
398
CHAPTER 16 The MIDI Studio: Equipment and Recording Procedures
■
■
■
■
■
■
■
■
you need to set your TV to a certain channel to see a station’s
programming, you need to set your MIDI sequencer’s track to
the same channel your controller is sending data on. It might
be easiest just to set the MIDI track input to your interface in
Omni mode, so the track will hear any channel that the controller is sending. You can set multiple MIDI tracks to Omni;
just record one track at a time.
The MIDI sequencer track output isn’t pointing at your soft
synth. You have to tell the MIDI track which synth to play.
The wrong sound bank was selected in a synth. You might have
specified a bank that contains no sounds. Or the MIDI mapping
on your controller is set so the notes aren’t playing the patches
you want them to play. Check the bank setting, patch settings,
and MIDI mapping.
In the sequencer’s options menu, under MIDI devices, your
MIDI interface isn’t selected. The sequencer doesn’t know where
to find the driver that communicates with your interface, so you
don’t hear anything. If your sequencer has a MIDI data indicator that is NOT flashing in response to your playing, maybe you
forgot to select the MIDI driver.
The synth’s audio-track volume, or the MIDI track’s key velocity,
is turned down. Disable any mute buttons and turn up the faders for those tracks. Check that no other tracks are soloed.
You started playing the MIDI track in the middle of a long note,
rather than at its beginning. A synth needs to receive a Note-On
command to play a note. If you start playing the sequence from
a point after the beginning of a note, the synth doesn’t receive
the note-on code, so it remains silent.
Your hardware synthesizer lost the program-change command.
Although you told the synth which patch to play, sometimes
this information is lost. Change the patch on the synth and
then set it back as it was. Or record the program change at the
beginning of the song.
The input to your sequencer track lost the setting for its MIDI
input device. Sometimes when you close a sequencer program
and reopen it, the MIDI track inputs might forget their settings.
Temporarily set the MIDI track input to another device, then set
it back as it was.
The MIDI driver is buggy. Download the latest update from the
interface manufacturer.
The MIDI Studio: Equipment and Recording Procedures CHAPTER 16
■
■
The synth audio output isn’t assigned to your interface’s stereo
output channels. In other words, the synth is making music but
not sending it anywhere.
Can you hear any audio playing through your monitors? If
not, the monitor power amp, or your mixer’s monitor controls,
might be off or turned down. Maybe the cables between your
interface and monitors aren’t plugged in. Seems obvious, but
it’s happened to me.
I hope these suggestions make your speakers play notes when you
press keys. The better you understand the MIDI/audio signal path,
the better you can find errors and fix them.
RECORDING WITH A KEYBOARD WORKSTATION
This is another way to record in a MIDI studio. Basically you follow
the steps for multitrack sequencer recording above, but do them in
the keyboard workstation rather than in a computer. The workstation’s manual tells how to do it.
You can access the keyboard’s effects menus to set up the effects that
are heard with a patch: hall reverb, chorus, flanging, echo, distortion,
and so on. Press the correct number on the numeric keypad to get to
the effects menus. (Note that these are built-in keyboard effects, not
outboard studio effects.)
Save the completed song (MIDI sequence) in multitrack form to the
keyboard’s internal memory or to a plug-in RAM card. If you’re satisfied with the final results, use a DAW to record the stereo output signals of the workstation. Then you can burn a CD or create an MP3
file from the WAVE file you just recorded on your hard drive.
In addition to these basic operations, you can:
■
■
■
■
■
■
■
Bounce tracks: copy one track’s performance to another track so
it will play another patch
Edit each note event
Create and copy patterns—for a drum or bass part, for example
Modify track and song parameters
Insert/delete/erase measures
Modify sounds and effects
Change the instrument (patch) that each track plays
399
400
CHAPTER 16 The MIDI Studio: Equipment and Recording Procedures
RECORDING WITH A DRUM MACHINE AND SYNTH
Suppose you want to record a drum pattern on a hardware drum
machine, and then you want to add a synth part using your computer.
First record the drum pattern. Here’s one suggested method:
1. On the drum machine, set the tempo, time signature, and pattern length in measures. In this example, the pattern will be 2
bars long.
2. Add a count-off (a few measures of clicks) at the beginning so
that overdubs made later can start at the correct time.
3. Start recording, and play the hi-hat key in time with the metronome beat.
4. At the end of 2 bars, the hi-hat pattern you tapped repeats over
and over (loops).
5. While this is happening, you can add a kick drum beat.
6. While the hi-hat and kick drum are looping, add a snare drum
beat, and so on.
7. Mix the recording by adjusting the faders or keys on the drum
machine for each instrument.
Next, you repeat the process for a different rhythmic pattern—say, a
drum fill—and store this as Pattern 2. Then develop other patterns.
Finally, you make a song by repeating patterns and chaining them
together as described in the drum machine’s instruction manual.
A song is a list of patterns in order.
Some musicians like to compose by programming a simple repeating drum groove first. While listening to this, they improvise a synth
part. After recording the synth part, they redo the drum part in detail,
adding hand claps, tom-tom fills, accents, and so on.
Finally, copy the audio signal of the drum pattern to a stereo track in
your computer recording program. Play back that track while overdubbing other parts.
USING EFFECTS
No matter how you record with MIDI, effects are an important part
of the mix. To keep the sound lively, try to vary the effects throughout the song, or use several types of effects in the mix.
The MIDI Studio: Equipment and Recording Procedures CHAPTER 16
For example, suppose you have a multitimbral synth, and you want
to add a different effect to each patch. Whether or not you can do
this depends on your synth. If it has a separate output for each
patch, you can use a different effect on each patch. But if your synth
has only a single output (mono or stereo) and you run it through an
effects device, the same effect is on all the patches.
If your song includes program changes (patch changes), you can have
the effects change when the patch changes. Set up a MIDI-driven
multi-effects processor so that each synth program change corresponds to the desired effect. When the synth program changes, the
effect changes also.
What if you want the effect, but not the synth patch, to change during
a mix? Reserve a track and channel just for effects program changes.
You don’t hear these program changes in your synth, but you do hear
the effects change. During a mixdown, it’s usually easier to change
effects automatically with your sequencer, rather than manually.
In a sample player, some samples might have reverberation or some
other effect already on them: the effect is part of the sampled sound.
Note that the sampled reverberation cuts off every time you play a
new note. Although this sounds unnatural, you can use it for special
effect.
Because effects are audio signals, audio recorders can record effects
but MIDI sequencers can’t. MIDI sequencers, however, can play
synths or patches that have effects built in.
Some recording software lets you convert a MIDI track to an audio
track, then apply audio effects to that track. Other software uses a
separate MIDI track and audio track for the same instrument patch.
You apply effects to the audio track.
MIDI effects (MFX) are non-audio processes applied to MIDI signals,
such as an arpeggiator, echo/delay, chord analyzer, quantize, transpose MIDI event filter, or velocity change. They are plug-ins in MIDI
tracks.
LOOP-BASED RECORDING
Let’s turn to a different aspect of computer MIDI recording. It’s possible to compose, record, and perform music entirely in software.
401
402
CHAPTER 16 The MIDI Studio: Equipment and Recording Procedures
You might start by creating a variety of loops or grooves, which are
constantly repeated rhythmic or musical patterns.
You can buy loops or make your own. You might create a loop by
recording a 4-measure sample of a drum beat or bass riff that you
play to a metronome or click track. Or use a handheld recorder to
sample some interesting sounds in the field. The sounds might create
a rhythm when repeated. If not, you can edit the timing of individual sounds until you get a rhythm you like. Soft synths are another
source for loops, and some are bundled with ready-to-use patterns.
Many recording programs include loop patterns. You can also create
loops from parts of tracks that you recorded.
Making Your Own Loops
Let’s make a 4-bar loop in 4/4 time using the editing function in a
DAW. Start with a musical pattern longer than 4 measures that was
made to a metronome or click track. Ideally the pattern has little or
no reverb so that the reverb won’t be cut off when you trim the loop.
Trim the start of the loop waveform just before beat 1, and trim the
end of the loop after measure 4-beat 4, and just before the next beat 1.
To avoid a click in the audio when the loop repeats, both trim points
should be at zero crossings where the waveform crosses the 0-volt line
(Figure 16.8)
Copy the trimmed 4-measure clip and paste it on the beat several
times to repeat it. Or right-click the clip and select Loop Clip. If you
hear no clicks and the rhythm is steady, export the 4-measure loop
Trim point
0 volts
FIGURE 16.8
Trim loops at zero crossings in the waveform.
The MIDI Studio: Equipment and Recording Procedures CHAPTER 16
to your loop library. Store your loop collection by tempo, style, and
type; for example, 120 bpm, jazz, drum set with brushes.
Once you have created several loops, you can drag and drop them
(or import them) into your DAW software. There, you can copy
and paste loops on the beat to repeat them. A SONAR groove clip
(explained next) can be repeated by click-dragging the end of the
clip to the right.
Types of Loops
There are five types of loop audio files based on their ability to have
their tempo changed:
■
■
■
■
■
A loop made from a standard digital audio file. It has a fixed
tempo, so you must build your song around that tempo.
A processed audio file. The tempo or pitch of the loop can be
changed by a time-stretch or pitch-shift algorithm in your audio
editing program. But if you need to change the tempo of a song,
you need to adjust all the loops you stretched.
A file with REX-based time-stretching. The transients in the
audio file are cut into slices, whose spacing depends on the
tempo. REX files follow tempo changes in your composition.
An acidized (RIFF) WAV file. Pitch and tempo information are
in the file header, and the audio is sliced at transients as with
REX files. Acidized files follow tempo and key changes. Slowing
down the loop by more than 15 or 20 bpm can add artifacts, so
it helps to start with a slow tempo of less than 120 bpm.
Cakewalk SONAR recording software includes Groove Clips that
have embedded tempo and root-note pitch. SONAR can stretch the
clips to match changes in tempo, or transpose clips to match the
project’s key. You repeat a Groove Clip by dragging its end to
the right in the track. Almost any audio clip can be converted to a
SONAR Groove Clip; just select the clip and press Ctrl-L.
Working with Loops
Once you’ve made several repetitions of the groove, you can add
effects, distortion, or pitch/tempo changes. You might put some echo
or reverb on a drum loop and vary the dry/wet mix with automation
as the loop plays. When the loops are complete, overdub audio and
MIDI tracks, such as soft-synth parts, percussion, vocals, and acoustic
403
404
CHAPTER 16 The MIDI Studio: Equipment and Recording Procedures
instruments. Import other loops and place them where you want
them in the song.
You can also create loops externally, then import them into a DAW
recording program. For example, compose a repetitive beat of drums,
synth, and samples in a beat box. Launch your DAW software, set up
a stereo track, and record the box output to that track as it plays the
beat. On other DAW tracks you can overdub vocals, doubled vocals,
rap sections, and harmony.
Repetitive loops can get boring, so edit or remove various notes
in the piano-roll view to add some variety. You might insert some
spaces here and there in the song to create breaks for solos or record
scratches.
Some loop programs offer groove quantizing, which transfers the
timing and dynamics from one groove to another. It allows humanlike variations in timing and key velocity.
Loop Libraries
Loop libraries are collections of sampled drum and bass beats that let
you loop (repeat), change tempo, etc. The beats come as MIDI files
and WAVE files. Some examples are M-Audio ProSessions, Hark Loops,
Cakewalk loop libraries, Sony Creative Software Artist Integrated loop
libraries, Sonic Foundry Acid loop libraries, GarageBand, Smart Loops,
FL Studio, Beatboy, Keyfax Twiddly Bits, Groove Monkee, FXpansion,
Discrete Drums, APO Multimedia Mix It, Multiloops Naked Drums,
Pocket Fuel RADS series, ILIO.com, Wizoo VST Drum Sessions, www
.bigfishaudio.com, and www.kellysmusicandcomputers.com/loop_
sample_libraries.htm. Do a Google search for “Loop Libraries” or
“ACID loops.” Complete looped rhythm tracks are available online for
purchase. Just add vocals.
Loop Creation Software
Many DAW recording programs include loop-creation software,
which is also available separately. For example, Propellerheads
Software (www.propellerheads.se) has a variety of programs to create
and modify loops. They also offer soft synths, drum machines, and
sequencers. Here are some of their products:
The MIDI Studio: Equipment and Recording Procedures CHAPTER 16
ReCycle is a stand-alone tool that starts with a loop and lets you
change its tempo and pitch, and replace and process sounds
within the loop. By detecting peaks in the waveform, ReCycle
automatically breaks a loop into parts or slices. A slice might
be a snare-drum hit, a kick-drum hit, or a kick-and-snare hit.
When a loop’s tempo is changed, the start point of each slice
moves in time so that the beats occur at the right time. (You
might need to touch this up manually.)
If the tempo is slowed down, ReCycle creates a decay after each
drum hit to fill in the gaps. You can delete slices, change their
length, attack, decay, and pitch; and add compression or EQ.
Then you can import the improved loop as an REX2 file into
an audio track in your sampler or sequencer program. There
you can control all aspects of the loop.
Reason is a group of synthesizers and samplers, a drum machine,
mixer, effects, pattern sequencer, and more (Figure 16.9).
It’s all-in-one and easy to learn. Reason can import and play
ReCycle’s REX files.
Reason Adapted is a lite version that is distributed as part of
some software bundles.
ReBirth is a software emulation of two analog bass-line synths
and two classic drum machines. Also included are delay, distortion, compression, and filtering. It integrates into sequencer
software and allows real-time audio streaming.
Reload is a utility that converts AKAI S1000 and S3000 formatted
media into formats (such as WAVE files) that can be used with
Reason, ReCycle, and other audio applications.
ReWire is a useful feature found in some loop programs. It transfers or streams audio data between two computer applications
in real time, almost like a cable. This allows programs to communicate with each other and synchronize together.
Other loop-based tools to compose and record music are listed below.
Ableton Live lets you compose with soft synths, record hi-resolution
multitrack audio, and play loops in a live performance. With Live
you can make stuttering drums, create arrangements, do MIDI loops,
and modify grooves, timing, pitch, volume, and effects.
Sony Creative Software’s ACID products let you select audio loops
from Windows Explorer, drag them into a recording program’s track
405
406
CHAPTER 16 The MIDI Studio: Equipment and Recording Procedures
FIGURE 16.9
The Thor synthesizer screen in Reason software.
view, and arrange them into multitrack projects. The tempo and key
of each loop are automatically matched to your project’s music in
real time, using the slicing technique described before. ACID uses
time-stretching algorithms to lengthen sounds when the tempo is
slowed down.
Other loop-intensive DAWs include Image Line Software FL Studio
(PC) and Apple GarageBand (with Apple Loops included). Some
excellent articles on loops were in the July 2004 and August 2002
issues of EQ Magazine and in the January 2007 issue of Recording
magazine.
CHAPTER 17
On-Location Recording of
Popular Music
Sooner or later you’ll want to record a band—maybe your own—
playing in a club or concert hall. Many bands want to be recorded
in concert because they feel that’s when they play best. Your job is to
capture that performance on a recorder and bring it back alive.
You can record pop-music concerts in several ways. Listed below are
a range of techniques from simple to elaborate. In general, sound
quality improves as the recording setup becomes more complex.
■
■
■
■
■
■
Record off the board (PA mixer) into a portable stereo recorder.
Record with two mics and a portable stereo recorder.
Record with a 4-tracker: Record with a stereo mic on tracks 1
and 2, while recording the PA mixer output on tracks 3 and 4.
Mix the tracks later.
Feed the PA mixer’s insert-send signals to a multitrack recorder.
Edit and mix the recording back in your studio.
Use a mic splitter on stage to feed the PA mixer, monitor mixer,
and recording mixer. Record to multitrack. Edit and mix the
recording back in your studio.
Do the multitrack recording in a van or truck.
We’ll describe each method so you can decide what’s best for you
and try it out.
RECORD OFF THE BOARD
Let’s start with the easiest technique: connect the PA mixing console
(board) to a portable digital recorder.
407
408
CHAPTER 17 On-Location Recording of Popular Music
Using cables with the appropriate mating connectors, connect the PA
mixer’s tape-out or 2-track-out connectors to your stereo recorder’s
line input(s). If your recorder’s line input is a stereo mini phone
jack, use an adapter cable: either two 1/4-inch phone plugs to a stereo mini phone, or two RCA connectors to a stereo mini phone. They
are available at Radio Shack and other electronic suppliers. Outside
the U.S., if your recorder’s line input is a stereo mini socket, use an
adapter cable: either two 6.35 mm jack plugs to a stereo mini jack, or
two phono connectors to a stereo mini jack.
The recorded mix might be poor with this method. Here’s why: The
sound mixer hears a combination of the band’s instruments, the
stage monitors, and the house speakers. So the board mix is intended
to augment the sound of the instruments and monitors on stage—
not to sound good by itself. For example, if the bass-guitar amp is
very loud on stage, it will be turned down in the mix that you record
off the board. Vocals will be turned up high in the mix because they
can’t be heard otherwise. Board mixes can sound good if there isn’t
much sound coming off the stage (as with acoustic groups), and the
venue is large or outdoors. It helps to monitor the board mix with
headphones or isolating earphones to hear what you’re recording.
RECORD WITH MICS AND A PORTABLE DIGITAL
RECORDER
Here’s another simple method. Try a stereo mic and a portable stereo recorder. The process is easy and the required gear costs as low as
$200. A stereo recording may not offer the sound quality of a professional multitrack recording. But if you record how the band sounds
from a seat in the audience, that may be good enough—especially if
the recording is just a demo, or is for the band members.
You can use a recorder with built-in microphones, or you can plug
two mics (or a stereo mic) into the recorder. You’ll pick up the group
as a whole from several feet away. The mics will capture not only
the musicians, but also the room reverb and background noise. You
could call it a “documentary” or “audio snapshot” recording.
As you record, what you hear with headphones or earphones is what
you get. There’s no mixing back in the studio.
If you place the mics a few feet from a folk group or jazz group without a PA system, the sound can be quite good. But most bands use
On-Location Recording of Popular Music CHAPTER 17
a PA. When you record the band you’re also recording the sound of
the PA speakers, so the mix or balance you get depends on the PA
engineer’s skill. The sound will be more distant and muddy than you
get with several close-up mics.
Gear
You’ll need a portable digital recorder, described in Chapter 4. A mic
stand is optional.
Headphones or earphones let you know whether the mics are working correctly, and let you hear what the mics are picking up. If the
band and PA are loud, it’s hard to hear what’s being recorded unless
you use isolating headphones (Remote Audio HN-7506) or isolating
earphones (Etymotic ER-4S, ER-4P, and ER-6i; Shure E3 and E4.)
Preparing for the Session
Before going on the road, install fresh batteries and clean the connectors with isopropyl alcohol or DeoxIT from Caig Labs (www.caig
.com). Do a trial recording to make sure everything works.
If possible, record the gig in a room where the audience is attentive
and the background noise is low. You might visit some potential
venues to check out the noise and acoustics. Avoid very live rooms
because they can make the recording muddy.
Be sure you have enough free space on your Flash memory card
before going on location. Listed below are approximate recording
times for a 1 GB card. Double these times for a 2 GB card.
24-bit/ 44.1 kHz stereo WAV—1.0 hour
16-bit/ 44.1 kHz stereo WAV—1.5 hours
256 kbps stereo MP3—9.0 hours
128 kbps stereo MP3—16.5 hours
The recorder might have a record-level switch labeled “manual” and
“auto.” Set it to manual to retain the dynamics of the performance. If
the switch is labeled “AGC” (Automatic Gain Control), set it to “off.”
At the Gig
Plug in some headphones and turn on Record Monitor mode. You’ll
hear the room acoustics and any background noise (audience, air
409
410
CHAPTER 17 On-Location Recording of Popular Music
PA
SPEAKER
MUSICIANS ON STAGE
top view
PA
SPEAKER
Alternative mic location
(see text)
RECORDER/MICS
top view
(see text)
FIGURE 17.1
Typical mic placement for a stereo recording of a pop group.
conditioning, traffic). Room noises that you wouldn’t otherwise
notice become obvious when you listen on headphones. Also listen
for buzzes, distortion, and crackles from bad cables or connections.
Where do you place the mics? The closer the mics are to the group,
the clearer and cleaner the sound is. In other words, close placement
captures more of the music and less of the room sound and background noise. Try to place the mics as close as possible to the stage
where you still pick up the house PA speakers well: about a stagewidth away from the stage (Figure 17.1). Keep the mics away from a
bar or other noise sources.
If there are dancers near the stage and the ceiling is low, you might
try boundary mics (such as two PZM®s) gaffer-taped to the ceiling or
mini mics hung from the ceiling.
To eliminate muddy room sound, you can record acoustic groups off
the PA mixer into a line input. Or try placing the recorder/mics on
stage (on a stool or mic stand) and record the sound of the monitor
speakers plus the band. Some bluegrass or old-time bands sing into
one microphone, so mount the recorder/mics next to that mic.
If the group plays in a circle (as in an outdoor jam or rehearsal), try
placing omnidirectional mics in the center. Or just walk around with
the mics while monitoring the mics over headphones. Find a spot
where you hear a good balance and put the mics there. Ask the musicians’ permission to record them during a break in the music.
On-Location Recording of Popular Music CHAPTER 17
Set the recording level so the meter reads about ⫺6 dB maximum.
That allows some headroom for surprises. Peak levels above 0 dB
result in distortion. Some portable recorders include a limiter which
prevents recording levels above 0 dB. Others have a clip or peak LED
that flashes when the recording level is too high.
Recorders have a mic gain switch or pad (attenuator) to prevent micpreamp distortion. If you have to set the recording-level control less
than 1/3 up to achieve a 0 dB recording level, use the low-gain setting
or switch in the pad.
If you’re using large-diaphragm condenser mics plugged into a phantom power supply, you might need to plug the output of the supply
into the recorder line input—rather than the mic input—to prevent
distortion.
A Recording Session with No PA
You might be able to record a small folk group or acoustic jazz group
on location without a PA system or audience. This gives you the freedom to improve the sound.
1. If the room is too live (reverberant), put up some packing
blankets, comforters, rugs, acoustic foam, or cushions.
2. Often a good-sounding spot for the musicians is near the center of a large room. Place the musicians around the stereo
mic pair where you want them to appear in the recording. For
example, you might place two singing guitarists on the left and
right, with bass in the center.
3. Experiment with microphone height to vary the vocal/guitar
balance. Try different miking distances to vary the amount of
ambience or room sound. A typical distance is three to six feet
(Figure 17.2).
4. As the musicians are playing and during playback, monitor the
mic signals with headphones. If some instruments or vocalists
are too quiet, move them closer to the mics, and vice versa,
until the balance sounds right.
5. While the band is rehearsing a loud song, make a trial recording. Set the recording level and mic gain as described earlier.
6. Record a tune. If someone makes a mistake, either record
another take of the entire tune, or record starting from a few
bars before the mistake, and edit the takes together later.
411
412
CHAPTER 17 On-Location Recording of Popular Music
SAX
DRUMS
PIANO
BASS
Players are located for
best recorded balance
STEREO MIC PAIR
FIGURE 17.2
A method of stereo miking a jazz group.
After the Gig
Back in the studio, connect the USB port in the recorder to the USB
port in your computer. The recorder shows up as a storage device on
your computer screen. Drag-and-drop the recorded sound files to the
computer’s hard drive for editing and CD burning. The files transfer
in a few minutes. Then the Flash memory card is empty and you’re
free to make more recordings.
Now you can edit the recording and adjust its tonal balance (equalize it) with DAW recording software or with Harmonic Balancer software (www.har-bal.com). You might cut a few dB around 300 Hz to
reduce boomy reverb and get a clearer recording. If your mics are
weak in the bass, compensate by boosting a few dB at 50 to 100 Hz.
Try out these tips and enjoy your recordings!
RECORD WITH A 4-TRACKER
This method is fairly simple and provides good sound. Put a stereo
mic or mic pair at the front-of-house (FOH) PA mixer position. Plug
the mic connectors into mic inputs 1 and 2 of a 4-track recorder–
mixer (or any size recorder–mixer). Connect the PA mixer’s tape outputs or 2-track outputs to line inputs 3 and 4 (Figure 17.3). Mix the
recording to stereo back in the studio.
The FOH mics pick up the band as the audience hears it: lots of
room reverb, lots of bass, but rather muddy or distant. The PA
mixer output sounds tight and clear, but typically is thin in the bass.
On-Location Recording of Popular Music CHAPTER 17
CARDIOID MICS
4–TRACK
RECORDER–MIXER
Tape out
or spare
main out
PA MIXER
FIGURE 17.3
Recording two mics and a PA mix on a four-track recorder–mixer.
Luckily, a mix of all four tracks can sound surprisingly good. Tracks 1
and 2 provide ambience and bass; tracks 3 and 4 provide definition
and clarity.
When you mix the four tracks, you might hear an echo because the
FOH mics pick up the band with a delay (caused by the sound travel
time from stage to mics). To remove the echo, import all the tracks
into digital recording software, and delay the PA mixer tracks by sliding them a little to the right. Align the waveforms of the mic tracks
and mixer tracks at big peaks.
CONNECT THE PA-MIXER INSERT SENDS TO
A MULTITRACK RECORDER
Now we get into professional techniques. This is an easy way to
record, and it offers very good sound quality with minimal equipment (Figure 17.4). Connect the mixer insert sends to the inputs
of a multitrack hard-drive recorder. Use the mixer gain trims to set
recording levels during the sound check. You don’t have to mix while
recording—instead, mix and monitor back in your studio. As for
drawbacks, you might have to ask the PA operator to adjust the gain
trims during the show to prevent recorder clipping.
Connections
This section describes how to connect a multitrack hard-drive
recorder to a mixer’s insert-send connectors.
413
414
CHAPTER 17 On-Location Recording of Popular Music
PA MIXER
MULTITRACK
RECORDER
Insert
sends*
Line
inputs
*Plug tip-sleeve connector in halfway
or wire tip/ring together
FIGURE 17.4
Connecting insert sends to a multitrack recorder provides great sound and easy setup.
TRIM
(TO SET
RECORD LEVEL)
MULTITRACK
RECORDER
IN
OUT
FADER
SEND
SIGNAL
IN
RETURN
EQ
PREAMP
INSERT JACK(S)
DIRECT OUT
JACK(S)
FIGURE 17.5
Simplified signal flow through part of a mixing console, showing insert and direct out.
In the console are several mic preamplifiers, one per mic, which
amplify the mic-level signal up to line level. For each mic channel,
this line-level signal typically appears at two connectors on the back
of the mixer: direct out and insert send. That’s where to connect to the
recorder inputs.
Usually the insert send is the best connector to use. Here’s why.
Typically the direct-out signal is post-fader (Figure 17.5). This means
the signal at the direct-out connector comes after the fader, so the
signal is affected by the fader (volume) settings. Any fader movements will show up on your recording, which is undesirable. It’s
better to connect recorder tracks to insert sends which are usually
pre-fader, pre-EQ. However, any changes the PA operator makes in
the trim settings during the show will affect your recording levels.
In some mixing consoles, the direct outs can be set to pre-fader. If so,
use those, because insert sends are often in use to connect compressors or gates.
On-Location Recording of Popular Music CHAPTER 17
SLEEVE (GROUND)
T
R
S
TO TIP-RING-SLEEVE
INSERT JACK
TIP
(SEND)
T
SHIELD
RING
(RETURN)
S
TO MACKIE
TIP-RING-SLEEVE
INSERT JACK
(INSERT ONLY TO FIRST CLICK)
WIRE TIP AND
RING TOGETHER
PHONE OR RCA
TO MULTITRACK IN
SLEEVE (GROUND)
TIP
(SEND)
PHONE OR RCA
TO MULTITRACK IN
TO RECORDER IN
TO SEND JACK
PHONE OR RCA
TO RETURN JACK
MONITOR INPUT ON MULTITRACK
TO RECORDER OUT
FIGURE 17.6
Three ways to wire cables based on the type of insert connector.
First, find out what kind of insert connectors the PA mixer has, and
what kind of input connectors your multitrack recorder has. Buy or
make some shielded cables (or a snake) that mate with those connectors. Figure 17.6 shows three ways to wire cables based on the
type of insert connector.
Some boards have a single TRS insert connector per channel, instead
of separate insert-send and insert-return connectors. Usually the tip
is send and the ring is return. In the TRS connector you plug into the
TRS insert, wire tip, and ring together, and also to the cable hot conductor (Figure 17.6, top). That way, the insert-send goes directly to
the insert-return. If nothing is connected to the insert return, no mic
signal goes through the mixer.
On some PA mixers with a TRS insert connector, you can use a TS
connector (Figure 17.6, middle). Plug it in halfway to the first click
so you don’t break the signal path—the mic signal still goes through
the PA mixer. If you plug in all the way to the second click, the signal doesn’t go through the PA mixer, just to the recorder. Cover the
415
416
CHAPTER 17 On-Location Recording of Popular Music
connections with a mixer case cover or board because someone could
bump into the mixer and dislodge a cable.
If the PA mixer has separate send and return insert connectors, connect the send to the recorder track input, and connect the return to the
recorder track output (Figure 17.6, bottom). If necessary, set your multitrack recorder to monitor the input analog signal so that the PA mixer
will receive a signal. Another option: Use an insert snake with TRS connectors at the mixer end. Carry some TRS-to-dual TS adapters to handle
mixers that have separate connectors for insert send and return.
The insert sends are balanced or unbalanced, and the same is true of
the recorder inputs. To connect balanced and unbalanced equipment
correctly, see the article “Sound System Interconnection” on the Rane
Web site, www.rane.com/note110.html.
You often encounter PA consoles where some insert sends are tied up
with signal processors. You must use the direct-out connectors of those
channels instead, which usually are post-fader (unless they can be
switched to pre-fader). Another option is to “Y” the insert send to your
recorder and to the processor input. Or, assign those mic channels to
unused groups (busses) and get your recording signals from there.
Some rack-mounted mic preamps have insert connectors. You could
plug the mic-snake XLRs into these preamps, then connect the preamp’s inserts to the PA mixer line inputs. That way, any gain changes
made on the PA console won’t affect your recording. Caution: Any
gain changes you make on the mic preamps will affect the PA levels.
If the PA mixer has a FireWire or USB port, simply connect the port
to a laptop that is running recording software. Set up the software to
recognize the mixer as its input/output device. The signal from each
mic channel goes to a separate track in the software.
What if you want to record several instruments on one track, such as a
drum mix? Assign all the drum mics to one or two output buses in the
PA mixer. Plug the bus-out insert connector to the recorder track input.
Use two buses for stereo. Adjust the faders to get a good drum mix.
Monitoring
To monitor the quality of the signals you’re recording, you generally
let the PA system be your monitor system. But you may want to set
On-Location Recording of Popular Music CHAPTER 17
up a monitor mix over isolating headphones or earphones so you
can hear more clearly.
Here is a suggested procedure. Connect all the recorder outputs to
unused line inputs in the PA mixer, or to a separate mixer. Use those
faders to set up a monitor mix. Assign them to an unused bus, and
monitor that bus with headphones/earphones. If you can spare only
a few inputs, plug in just one track at a time to check its sound quality. Listen closely for any hum, noise, or distortion.
Setting Levels
Set recording levels with the PA mixer’s gain-trim or input-attenuator
knobs. This affects the levels in the PA mix, so be sure to discuss your
trim adjustment in advance with the PA mixer operator. If you turn
down an input trim, the PA operator must compensate by turning
up that channel’s fader and monitor send.
As we said, if the PA operator changes the input trim during the
show, these changes will show up on your recording.
Set recording levels before the concert during the sound check
(if any!). It is better to set the levels a little too low than too high
because during mixdown you can reduce noise but not distortion. A
suggested starting level is ⫺10 dB, which allows for surprises. Don’t
exceed 0 dB because the signal will distort. Also, if you set the recording level conservatively, you’re less likely to change the gain trims
during the performance. You don’t want to hassle the PA operator.
Some recording engineers run each insert-send signal through a
potentiometer to set the recording level. By adjusting levels on a
rack-mounted panel of potentiometers, you don’t have to ask the PA
operator to change the gain trims for you. Of course, if you’re operating the PA console, you can set the gain trims yourself.
If you have a spare recorder, record a safety copy on it at the same
time. This provides a backup in case one recorder fails.
SPLITTING THE MIC SIGNALS
Now let’s consider a different way to make a multitrack recording. Plug
each mic into a mic splitter, which sends the mic signal to two destinations: the PA mixer and recording mixer. This gives you independent
417
418
CHAPTER 17 On-Location Recording of Popular Music
control of each microphone. The splitter has one XLR input and two
or three XLR outputs per mic. The third output on some splitters
goes to a monitor mixer. You can use transformer-isolated splitters or
Y-cable splitters.
Splitters are made by such companies as Pro Co, Radial, Whirlwind,
Rapco, Klark Technik, and Wireworks.
As shown in Figure 17.7, connect the outputs from all the splitter
channels to the PA snake and to your recording snake. Connect the
recording snake to your recording mixer (or audio interface) mic
inputs. Connect the recording mixer’s insert sends to a multitrack
recorder, or connect the interface USB/FireWire cable to a laptop
computer. When the recording is done, you can edit and mix the
recording back in your studio.
Let’s explain the advantages of splitting the mic signals. You use your
own mic preamps, so you aren’t dependent on the quality of the PA
console mic preamps. Also, you aren’t hassling the operator about
adjusting gain trims. Each mix engineer can work without interfering
with the other. The FOH engineer can change trims, level, or EQ, and
it will have no effect on the signals going to the recording engineer.
Another plus: A splitter provides consistent, unprocessed recordings
RECORDING MIXER
STAGE
BOX
MULTITRACK
RECORDER
CABLES
Insert
sends
MIC SPLITTER
Track
inputs
SNAKE
MICS
DIRECT BOX
AMPLIFIERS/
SPEAKERS
SNAKE
STAGE
BOX
PA MIXER
FIGURE 17.7
Splitting the mic signals to the recording mixer and PA mixer.
On-Location Recording of Popular Music CHAPTER 17
of the mic signals. This consistency makes it easy to edit between different performances. What’s more, splitters let you use mic preamps
on stage if you wish. That way, the cable from each mic to its preamp
is short, which reduces hum and radio-frequency interference.
Using Splitters
To use a splitter or multichannel splitter, plug each mic into a splitter
input. Decide which mixer you want to supply phantom power (usually the PA mixer). Connect the splitter’s direct outputs to that mixer’s snake. Connect one set of isolated outputs to the recording snake,
and another set to the monitor snake (if used). Plug the snakes into
the mixers. Figure 17.8 shows splitter connections to three mixers.
Splitters have a ground-lift switch on each output channel. This
switch connects or disconnects (floats) the cable shield from pin 1
of the XLR connector. When the ground-lift switches are set correctly,
you should get no ground loops and their resulting buzzes.
How do you set the ground-lift switches?
1. First, turn off phantom power in each console. Turn down all
the faders.
FIGURE 17.8
A transformer-isolated microphone splitter.
419
420
CHAPTER 17 On-Location Recording of Popular Music
2. Make sure the direct feed’s ground-lift switches (if any) are set
to ground, not lift. Otherwise phantom power won’t work.
3. Go to the mixer connected to the direct feed. Turn on the
mixer, switch on phantom power, and bring up each fader to
listen for a signal.
4. On the splitter, find the ground-lift switches for the recording-mixer feed. Set them to the position where you monitor
the least hum and buzz at the recording mixer. In XLR Y-cable
splitters, solder a ground-lift switch between the shield and pin
1 in the non-direct feeds.
5. Repeat step 4 for the monitor mixer.
MULTITRACK RECORDING IN A TRUCK
Here’s the ultimate setup. Each mic is split three ways to feed the
stage boxes for the recording, reinforcement, and monitor consoles.
A long snake is run to a recording truck or van parked outside the
concert hall or club. In the truck, the snake connects to a mixing console plugged into a multitrack recorder. The interior of the truck is
acoustically treated, and powered Nearfield monitor speakers allow
accurate monitoring of the signal. Designing a recording truck is a
subject in itself, and is beyond the scope of this book.
The truck contains the console, monitors, and recorders, so you don’t
need to cart them into the venue. This saves setup time. Also, a truck
provides a quiet, controlled monitoring environment.
There you have a full choice of methods for recording live. Check out
all the options, and you’ll find a system that works well for your style
of recording.
PREPARING FOR THE SESSION
Ready to record a live gig? The recording will go a lot smoother if
you plan what you’re going to do. So sit down, grab a pen, and make
some lists and diagrams as described here. We’ll go over the steps to
plan a recording.
Preproduction Meeting
Call or meet with the sound-reinforcement company and the production company putting on the event. Find out the date of the event,
On-Location Recording of Popular Music CHAPTER 17
location, phone numbers, and e-mails of everyone involved; when
the job starts; when you can get into the hall; when the second set
starts; and other pertinent information. Decide who will provide the
split, which system will be plugged in first, second, and so on. Draw
block diagrams for the audio system and communications (comm)
system. Determine who will provide the comm headphones.
If you’re using a mic splitter, work out the splitter feeds. The mixer
getting the direct side of the split provides phantom power for condenser mics that aren’t powered on stage. If the house system has
been in use for a long time, give them the direct side of the split.
Overloud stage monitors can ruin a recording, so work with the
sound-reinforcement people toward a workable compromise. Ask
them to start with the monitors quiet because the musicians always
want them turned up louder.
Make copies of the meeting notes for all participants. Don’t leave things
unresolved. Know who is responsible for supplying what equipment.
Site Survey
If possible, visit the recording site in advance and go through the following checklist:
■
■
■
■
■
■
■
Check the AC power to make sure the voltage is adequate, the
third pin is grounded, and the waveform is clean.
Listen for ambient noises: ice machines, coolers, 400-Hz generators, heating pipes, air conditioning, nearby discos, etc. Try to
have these noise sources under control by the day of the concert.
Sketch dimensions of all rooms related to the job. Estimate distances for cable runs.
Turn on the sound-reinforcement system to see if it functions
okay by itself (no hum, and so on). Turn the lighting on at various levels with the sound system on. Listen for buzzes. Try to
correct any problem so that you don’t document bad PA sound
on your recording.
Determine locations for any audience/ambience mics. Keep
them away from air-conditioning ducts and noisy machinery.
Plan your cable runs from stage to recording mixer.
If you plan to hang mic cables, feel the supports for vibration.
You may need microphone shock mounts. If there’s a breeze in
the room, plan on taking windscreens.
421
422
CHAPTER 17 On-Location Recording of Popular Music
■
■
■
Make a file on each recording venue including the dimensions
and the location of the circuit breakers.
Determine where the control room will be. Find out what surrounds it—any noisy machinery?
Visit the site when a crowd is there to see where there may be
traffic problems.
Mic List
Now write down all the instruments and vocals in the band. If you
want to put several mics on the drum kit, list each drum that you
want to mike. As for keyboards, decide whether you want to record
off each keyboard’s output, or off the keyboard mixer (if any).
Next, write down the mic or direct box you want to use on each
instrument (see Table 17.1).
Table 17.1
Mic List (Example)
Input
Instrument
Mic
1
Bass
DI
2
Kick
AKG D112
3
Snare
Shure Beta 57
4
Hi-hat
Crown CM-700
5
Small rack tom
Shure SM57
6
Large rack tom
Shure SM57
7
Small floor tom
Senn. MD421
8
Large floor tom
Senn. MD421
9
Overhead L
AT 4051A
10
Overhead R
AT 4051A
11
Lead guitar
Shure SM57
12
Rhythm guitar
Shure SM57
13
Keyboard mixer
DI
14
Lead vocal
Beyer M88
15
Harmony vocal
Crown CM-311A
On-Location Recording of Popular Music CHAPTER 17
Make copies of this mic list. At the gig, you’ll place one list by the
stage box, and the other by each mixing console. The list will act as a
guide to keep things organized.
Track Sheet
Next, decide what will go on each track of your multitrack recorder. If
you have enough tracks, your job is easy: Just assign each instrument
or vocal to its own track: bass to track 1, kick to track 2, and so on. If
you need to assign several instruments or vocals to the same track, set
up a submix as described earlier. Table 17.2 shows a sample track list.
Block Diagram
Now that your track assignments are planned, you can figure out
what equipment you’ll need. Draw a block diagram of your recording setup from input to output (such as Figure 17.9). Include mics,
mic cables, mic stands and booms, DI boxes, insert cables, multitrack
recorder(s), outlet strip and extension cord, and recording media.
You might bring your own mixer and snake, or use those from the
house system. On your diagram, label the cable connectors on each
end so you’ll know what kinds of cables to bring. It’s a good idea to
keep a file of system block diagrams for various recording venues.
In Figure 17.9, the block diagram shows a typical recording method:
feeding FOH console insert jacks to a multitrack recorder. We’ll use
this example in the rest of this chapter.
Table 17.2
Track Sheet (Example)
Track
Instrument
1
Drum mix L
2
Drum mix R
3
Bass
4
Lead guitar
5
Rhythm guitar
6
Keys mix
7
Lead vocal
8
Harmony vocal
423
424
CHAPTER 17 On-Location Recording of Popular Music
MICS
MIC STANDS
GUITAR CORD
MIC CABLES
SNAKE
DI
INSERT
SNAKE
MULTITRACK
RECORDER
(Hard drive recorder or
interface feeding a laptop)
PA MIXER
HARD DRIVES
Insert
sends
Line
inputs
OUTLET STRIP
TO EXT. CORD
FIGURE 17.9
Sample block diagram.
Equipment List
Generate a list of recording equipment from your block diagram.
Based on Figure 17.9, you’d need the following recording gear:
■
■
■
■
■
Direct boxes, mics, mic cables, mic stands, and booms (unless
these are part of the house system)
Insert cables (for example, 8 stereo phone plugs to 8 RCA
connectors)
Multitrack recorder
Outlet strip and extension cord
Recording medium (bring enough for the duration of the gig)
Don’t forget the incidentals: a cleaning tape, pen, notebook, flashlight, guitar picks, heavy-duty guitar cords, drum keys, spare tape, mic
pop filters, gaffer’s tape, guitar tuner, ear plugs, audio-connector adapters, audio cable ground-lift adapters, in-line pads, in-line polarity
On-Location Recording of Popular Music CHAPTER 17
reversers, spare cables, gooseneck lights for the console, spare batteries, water, food—and aspirin!
Bring a tool kit with screwdrivers, pliers, soldering iron and solder,
AC-outlet checkers, fuses, a pocket radio to listen for interference,
ferrite beads of various sizes for RFI suppression, canned air to shoot
out dirt, cotton swabs and pipe cleaners, and DeoxIT from Caig Labs
to remove oxide from connectors.
Check off each item on the list as you pack it. After the gig, you can
check the list to see whether you reclaimed all your gear.
PREPARING FOR EASIER SETUP
You want to make your setup as fast and easy as possible. Here are
some tips to help this process.
Protective Cases
Mount your console and recorders in protective carrying cases. Install
casters or swivel wheels under racks and carrying cases so you can
roll them in. Rolling is so much easier than lifting and carrying. You
might permanently install the multitrack recorders in SKB carrying
cases that act as racks. When a remote job comes up, just grab them
and go.
A very helpful item is a dolly, wheeled cart, or hand truck to transport heavy equipment into the venue. Consider getting some lightweight tubular carts. Being collapsible, they store easily in your car
or truck. One maker of equipment carts is www.kart-a-bag.com.
Pack mics, headphones, and other small pieces in cloth bags, trunks,
or milk crates.
Caution: Keep tapes and hard drives separated from magnets, such
as in headphones, monitor speakers, and dynamic mics. (However,
tape erasure is much less of a problem with digital media than it is
with analog tapes.)
You might want to build a mic container: a big box full of foam rubber with cutouts for all the mics. Or construct a wheeled cabinet with
drawers for mics, DIs, and speaker cables.
425
426
CHAPTER 17 On-Location Recording of Popular Music
In an article in the September–October 1985 issue of db Magazine,
remote recording engineer Ron Streicher offered these suggestions:
“Especially for international travel, make sure your documentation is
up to date and matches the equipment you’re carrying. Make a list of
everything you take: all the details, such as each pencil, razor blade,
connector, etc. Also make sure your insurance is up to date. You need
insurance for en route as well as at the destination.
“I organize my cases so I know where every item is. They’re ready to
go anytime and make setup much faster. The cables are packed with
their associated equipment, not in a cables case. I check everything
coming and going, and try to have 100% redundancy, such as a small
mixer to substitute for the large console.”
Mic Mounts
If you’ll be recording a singer/guitarist, take a short mic mount that
clamps onto the singer’s mic stand. Put the guitar mic in the mount.
Also bring some short mounts or mini mics to clamp onto drum
rims. By using these mounts, you eliminate the weight and clutter of
several mic stands.
Some examples of short mounts are the Mic-Eze units by Ac-cetera
(www.ac-cetera.com). They have standard 5/8–27 threads and mic
clamps that either spring shut or are screw-tightened. Flex-Eze is
two clamps joined by a short gooseneck and Min-Eze is two clamps
joined by a swivel.
Snakes and Cables
You can store mic cables on a cable spool, available in the electrical department of a hardware store. Wrap one mic cable around the
spool, plug it into the next cable and wrap it, etc.
A snake can be wrapped around a large reel or can be coiled in a
trunk. Commercial snake reels are made by such companies as
Whirlwind (www.whirlwindusa.com), ProCo (www.procosound.com),
and Hannay (www.hannay.com).
Use wire ties to join cables that you normally run together, such as
PA sends and returns.
On-Location Recording of Popular Music CHAPTER 17
Snake hookup is quicker if the snake has a multipin twist-lock connector (such as Whirlwind W1 or W2). This connector plugs into a
mating connector that divides into several male XLRs. Those XLRs
plug into the mixing console. Leave the XLRs in the console carrying
case. You’ll find that the snake is easier to handle without the XLR
pigtails on it.
For a clean, rapid hookup of drum mics, put a small snake near
the drum kit, and run it to the main stage box. Or, put the main
stage box by the drum kit. Snakes are made by such companies as
Whirlwind, ProCo, and Horizon (www.horizonmusic.com).
Check that all your mic cables are wired in the same polarity: pin 2
hot on both ends.
You might want to use 3-conductor shielded mic cables. Connect
the shield to ground only at the male XLR end. Also use cables with
100% shielding. Those measures enhance the shielding capability of
the shield and reduce pickup of lighting buzzes.
In XLR-type cable connectors, don’t wire pin 1 to the shell, or you
may get ground loops when the shell contacts a metallic surface.
Label all your cables on both ends according to what they plug into;
for example, DSP-9 effects in, track 12 out, power amp in, snake aux
2 out. Or you might prefer to number the cables near their connectors. Cover these labels with clear heat-shrink tubing.
Label both ends of each mic cable with the cable length. Put a drop
of glue on each connector screw to temporarily lock it in place.
Rack Wiring
You can speed the console wiring by using a small snake between the
rack and the console, and between the multitrack recorders and the
console. When packing, plug the snakes into the rack gear and multitracks, and coil the snake inside the rack and the multitrack carrying
case. In other words, have all your equipment pre-wired. At the gig,
pull out a bundled harness and plug it into the console jacks.
You might be feeding your multitrack from the insert jacks of the
house console. If so, use a snake with TRS stereo phone plugs at the
427
428
CHAPTER 17 On-Location Recording of Popular Music
console end. Carry some TRS-to-dual-mono phone adapters to handle consoles that have separate jacks for insert send and return.
Some engineers prefer to make a clearly marked interface panel on
the rear of the racks and plug into the panels. This is easier than trying to find the right connectors on each piece of equipment.
Small bands might get by with all their equipment in a single, tall
rack. Mount a small mixer on top, wired to effects in the middle,
with a coiled snake and power cord on the bottom.
Small snakes for rack and multitrack connections are made by Hosa
(www.hosatech.com), Horizon, and ProCo, among others.
Other Tips
Here are some more helpful hints for successful on-location
recordings.
■
■
■
■
■
■
■
■
Record several gigs and pick the best performances to use in an
album.
Plan to use a talk-back mic from the board to the stage monitor
speakers during sound checks. You might bring a small instrument amp for talkback so that you can always be heard.
Hook up and use unfamiliar equipment before going on the
road. Don’t experiment on the job!
Consider recording with redundant (double) systems so you
have a backup if one fails.
Walkie-talkies are okay for pre-show use, but don’t use them
during the performance because they cause RF interference. Use
hard-wired communications headsets. Assistants can relay messages to and from the stage crew while you’re mixing.
During short set changes, use a laptop computer Local Area
Network to show what set changes and mic-layout changes are
coming up next; transmit this information to the monitor mixer
and sound-reinforcement mixer.
Don’t put tapes or hard drives through airport X-ray machines
because the transformer in these machines isn’t always well
shielded. Have the tapes or hard drives inspected by hand.
Hand-carry your mics on airplanes. Arrange to load and unload
your own freight containers, rather than trusting them to airline
freight loaders. Expect delays here and at security checkpoints.
On-Location Recording of Popular Music CHAPTER 17
■
■
■
■
■
■
Get a public-liability insurance policy to protect yourself against
lawsuits.
Call the venue and ask directions to the load-in door. Make
sure that someone will be there at setup time to let you in. Ask
the custodian not to lock the circuit-breaker box the day of the
recording.
A few days before the session, check out the parking situation.
Just before you go, check out all your equipment to make sure
it’s working.
Arrive several hours ahead of time for parking and setup. Expect
failures—there’s always something going wrong, something
unexpected. Allow 50% more time for troubleshooting than
you think you’ll need. Have backup plans if equipment fails.
In general, plan everything in advance so you can relax at the
gig and have fun!
AT THE SESSION: SETUP
Okay! You’ve arrived at the venue. After parking, offload your gear to
a holding area, rather than onstage, because gear on stage will likely
need to be moved.
Learn the names of the house sound-crew members, and be friendly.
These people can be your assets or your enemies. Think before you
comment to them! Try to remain in the background and don’t interfere with their normal way of doing things (for example, take the
secondary side of the split).
Power Distribution System
At the job, you need to take special precautions with power distribution, interconnecting multiple sound systems, and electric guitar
grounding.
Consider buying, renting, or making your own single-phase power
distribution system (distro). It will greatly reduce ground loops and
increase reliability. One source of AC power distribution equipment
is www.furmansound.com. Figure 17.10 shows a suggested AC power
distribution system.
The amp rating of the distro’s main breaker box should exceed the
current drain of all the equipment that will be plugged into the
distro system.
429
430
CHAPTER 17 On-Location Recording of Popular Music
FIGURE 17.10
An AC power distribution system for a touring sound system.
Power Source
If you’re using a remote truck, find a source of power that can handle the truck’s power requirements, usually at a breaker panel. Some
newer clubs have separate breaker boxes for sound, lights, and a
remote truck. Find out whether you’ll need a union electrician to
make those connections. Label your breakers.
Check that your AC power source isn’t shared with lighting dimmers or
heavy machinery; these devices can cause noises or buzzes in the audio.
The industry-standard power connector for high-current applications
is the Cam-lok, a large cylindrical connector. Male and female Camloks join together and lock when you twist the connector ring. Distro
systems and power cables with Cam-lok connectors can be rented
from rental houses for film, lighting, electrical equipment, or entertainment equipment. One such rental house is Mole-Richardson at
www.mole.com.
Use an adapter from Cam-lok to bare wires. Pull the panel off the
breaker box, insert the bare wires, and connect the Cam-lok to your
truck’s power.
Caution: Have an electrician do the wiring if you don’t know what
you’re doing. A union electrician might be required anyway. Some
breaker boxes have Cam-loks already built in.
On-Location Recording of Popular Music CHAPTER 17
To reduce ground-loop problems, get on the same power that the
house sound system is using. From that point, run your distribution
system, or at least run one or two thick (14- or 16-gauge) extension
cords, to your recording system. These cords may need to be 100 to
200 feet long. Plug AC outlet strips into the extension cord; then
plug all your equipment into the outlet strips.
If your recording system is one or two multitrack recorders that will
connect to the FOH console, simply plug into the same outlet strip
that the FOH console is using.
Measure the AC line voltage. If the AC voltage varies widely, use a
line voltage regulator (power conditioner) for your recording equipment. If the AC power is noisy, you might need a power isolation
transformer.
Check AC power on stage with a circuit checker. Are grounded outlets
actually grounded? Is there low resistance to ground? Are the outlets
the correct polarity? There should be a substantial voltage between
hot and ground, and no voltage between neutral and ground.
Some recording companies have a gasoline-powered generator ready
to use if the house power fails. If there are a lot of lighting and dimmer racks at the gig, you might want to put the truck on a generator
to keep it isolated from the lighting power.
Interconnecting Multiple Sound Systems
If you hear hum or buzz when the systems are connected, first make
sure that the signal source is clean. You might be hearing a broken
snake shield or an unused bass-guitar input.
If hum persists, experiment with flipping the ground-lift switches
on the splitter and on the direct boxes. On some jobs you need to
lift almost every ground. On others you need to tie all the grounds.
The correct ground-lift setting can change from day to day because
of a change in the lighting. Expect to do some trial-and-error
adjustments.
If the house system has serious hum and buzz problems, offer help.
You can hear buzzes in your quiet truck that they can’t hear over the
main system with noise in the background.
431
432
CHAPTER 17 On-Location Recording of Popular Music
Often, a radio station or video crew will take an audio feed from
your mixing console. In this case, you can prevent a hum problem
by using a console with transformer-isolated inputs and outputs. Or
you can use a 1:1 audio isolation transformer between the console
and the feeds. For best isolation, use a distribution amp with several transformer-isolated feeds. Lift the cable shield at the input of
the system you’re feeding. Some excellent isolation transformers are
made by Jensen (www.jensen-transformers.com).
Connections
After unpacking, place one mic list by the stage box so you know
what to plug in where. Place a duplicate list by each console. Attach
a strip of white tape just below the mixer faders. Use this strip to
write down the instrument that each fader affects.
Based on the mic list you wrote, you might plug the bass DI into
snake input 1, plug the kick mic into snake input 2, and so on. Label
fader 1 “BASS,” label fader 2 “KICK,” etc. Also plug in equipment
cables according to your block diagram.
Have an extra microphone and cable offstage ready to use if a mic fails.
Don’t unplug mics plugged into phantom power because this will
make a popping noise in the sound-reinforcement system.
Running Cables
To reduce hum pickup and ground-loop problems associated with
cable connectors, try to use a single mic cable between each mic and
its snake-box connector.
Avoid bundling mic cables, line-level cables, and power cables
together. If you must cross mic cables and power cables, do so at
right angles and space them vertically.
Plug each mic cable into the stage box; then run the cable out to the
each mic and plug it in. This leaves less of a mess at the stage box.
Leave the excess cable at each mic stand so you can move the mics.
Don’t tape the mic cables down until the musicians are settled.
It’s important that audience members don’t trip over your cables.
In high-traffic areas, cover cables with rubber floor mats or cable
On-Location Recording of Popular Music CHAPTER 17
crossovers (metal ramps). At least tape them down with gaffer’s tape
pressed lengthwise onto the cables.
It helps to set up a closed-circuit TV camera and TV monitor to see
what’s happening on stage. You need to know when mics get moved
accidentally, or when singers use the wrong mic, etc.
Recording-Console Setup
Here’s a suggested procedure for setting up the recording system
efficiently:
1. If the console is set up in a dressing room or locker room, add
some acoustic absorption to deaden the room reflections. You
might bring a carpet for the floor, plus acoustic foam or moving blankets for the walls.
2. Turn up the recording monitor system and verify that it is clean.
3. Plug in one mic at a time and monitor it to check for hums
and buzzes. Troubleshooting is easier if you listen to each mic
as you connect it, rather than plugging them all in, and trying
to find a hum or buzz.
4. Check and clean up one system at a time: first the sound-reinforcement system, then the stage-monitor system, then the
recording system. Again, this makes troubleshooting easier
because you have only one system to troubleshoot.
5. Use as many designation strips as you need for complex consoles. Label the input faders bottom and top. Also label the
monitor-mix knobs and the meters.
6. Monitor the reverb returns (if any) and check for a clean
signal.
7. Make a short test recording and listen to the playback.
8. Verify that left and right channels are correct, and that the panpot action isn’t reversed audibly.
9. If you’re setting up a separate recording monitor mix, do a preliminary pan-pot setup. Panning similar instruments to different locations helps you identify them.
MIC TECHNIQUES
Usually the miking is left up to the sound-reinforcement company.
But there are some mic-related problems you should know about,
433
434
CHAPTER 17 On-Location Recording of Popular Music
such as feedback, leakage, room acoustics, and noise. Here are some
ways to control these problems:
■
■
■
■
■
■
Use directional microphones, such as cardioid, supercardioid,
or hypercardioid. These mics pick up less feedback, leakage, and
noise than omnidirectional mics at the same miking distance.
With vocal mics, aim the null of the polar pattern at the floor
monitors. The null (area of least pickup) of a cardioid is at the
rear of the mic—180 degrees off-axis. The null of a supercardioid is 125 degrees off-axis; hypercardioid is 110 degrees.
To reduce breath pops with vocal mics, be sure to use foam pop
filters. Allow a little spacing between the pop filter and the mic
grille. It also helps to switch in a low-cut filter (100-Hz highpass
filter).
Mike close. Place each mic within a few inches of its instrument. Ask vocalists to sing with lips touching the mic’s foam
pop filter.
Use direct boxes. Bass guitar and electric guitar can be recorded
direct to eliminate leakage and noise in their signals. However,
you might prefer the sound of a miked guitar amp. You could
record the guitar direct from its effects boxes. Then use a guitaramp emulator during mixdown. Note that sequencers and
some keyboards have high-level outputs, so their DI boxes need
transformers that can handle line level.
Use contact pickups. On acoustic guitar, acoustic bass, and
violin, you can avoid leakage by using a contact pickup. Such
a pickup is sensitive only to the instrument’s vibration, not so
much to sound waves. The sound of a pickup isn’t as natural as
a microphone, but a pickup may be your only choice. Consider
using both a pickup and a microphone on the instrument. Feed
the pickup to the house and monitor speakers; feed the mic and
pickup to the recording mixer.
When you’re recording a band that has been on tour, should you use
their PA mics or your own mics? In general, go with their mics. The
artists and PA company have been using their mics for a while and
may not want to change anything. Most mics currently used in PA
are good quality anyway, unless they are dirty or defective.
If you’re not happy with their choice, you could add your own instrument mics. Let the PA people listen to the sound in the recording
On-Location Recording of Popular Music CHAPTER 17
truck, or in headphones. If it sounds bad because of their mic choice,
ask “Would it be okay if we tried a different mic (or mic placement)?” Usually it’s all right with them—it’s a team effort.
Electric Guitar Grounding
While setting up mics, you need to be aware of a safety issue with
the electric guitar. Electric-guitar players can receive a shock when
they touch their guitar and a mic simultaneously. This occurs when
the guitar amp is plugged into an electrical outlet on stage, and the
mixing console (to which the mics are grounded) is plugged into a
separate outlet across the room. If you’re not using a power distro,
these two power points may be at widely different ground voltages.
So a current can flow between the grounded mic housing and the
grounded guitar strings.
Caution: Electric guitar shock is especially dangerous when the guitar amp and the console are on different phases of the AC mains.
It helps to power all instrument amps and audio gear from the same
AC distribution outlets. If you lack a power distro, run a heavy extension cord from a stage outlet back to the mixing console (or vice
versa). Plug all the power-cord ground pins into grounded outlets.
That way, you prevent shocks and hum at the same time.
If you’re picking up the electric guitar direct, use a transformerisolated direct box and set the ground-lift switch to the minimumhum position.
Using a neon tester or voltmeter, measure the voltage between the
electric-guitar strings and the metal grille of the microphones. If
there is a voltage, flip the AC polarity switch on the amp (if any).
Use foam windscreens for additional protection against shocks.
Audience Microphones
If you have enough mic inputs, you can use two audience mics to
pick up the room acoustics and audience sounds. This helps the
recording to sound “live.” Without audience mics, the recording may
sound too dry, as if it were done in a studio.
One easy method is to aim two cardioid mics at the audience. Put them
on tall mic stands, on the stage floor, and on either side of the stage. If
those mic stands must not be seen, try hanging mics (Figure 17.11).
435
436
CHAPTER 17 On-Location Recording of Popular Music
SIDE VIEW OF MIC PLACEMENT
HANGING MICS
F.O.H. MICS
STAND
MICS
STAGE
AUDIENCE
FIGURE 17.11
Some audience miking techniques.
Some engineers pick up the audience and hall with two mics placed
at the FOH mix position. That’s an easy, effective way to capture
the crowd. But the signal from these mics is delayed relative to the
on-stage mics. If the FOH mix position is far from the stage (say, 50
feet or more), this delay will cause an echo when the audience mics
are mixed with the stage mics. Prevent this by placing the audience
mics fairly near the stage. Or, during computer mixdown, move the
FOH mic tracks to the left, earlier in time so that large peaks in the
FOH tracks’ waveform coincide with large peaks in on-stage tracks.
If you’re using the audience mics only to capture applause, the delay
isn’t an issue.
What if you don’t have enough tracks for the audience mics? Record
them on a 2-track recorder. Load this recording into your DAW along
with the stage-mic mix. Align the two recordings in time as described
above.
If the audience mics are run through the FOH mixer, leave the audience mics unassigned in that mixer to prevent feedback.
To get more isolation from the house speakers in the audience mics,
use several mics hung close to the audience. Some engineers put up
four audience mics maximum; some use eight to ten. Use directional
mics and aim the rear null at the house speakers.
Another option is not miking the audience, or not using the audience
tracks. Instead, during mixdown, you could simulate an audience with
an audience-reaction CD. Simulate room reverb with an effects unit.
On-Location Recording of Popular Music CHAPTER 17
SOUND CHECK AND RECORDING
Now that the mics are set up, you might have time for a sound
check. That’s when you set recording levels as described earlier in the
section Setting Levels.
A few minutes before the band starts playing, start recording. Keep a
close eye on recording levels. If a track is going into the red, slowly
turn down its input trim and note the recorder-counter time where
this change occurred.
Caution: If you’re recording off the FOH mixer, turning down its
input trim will affect the house levels. The FOH mixer operator will
need to turn up the corresponding monitor send and channel fader.
This is a touchy situation that demands cooperation. Ideally, you
set enough headroom during the sound check so you won’t have
to change levels. But be sure the house mixer operator knows in
advance that you might need to make changes. Ask the operator
whether she wants to adjust the gain trims for you, so she can adjust
corresponding levels at the same time. Thank the operator for helping you get a good recording.
If you’re recording with a splitter and mic preamps on stage, assign
someone to watch the levels and adjust them during the concert.
Preamps with meters allow more precise level setting than preamps
with clip LEDs.
Keep a track sheet and log as you record. For each song in the set list,
note the recorder-counter time when the song starts, or press the Set
Locate button on your multitrack recorder. Later, during mixdown,
you can go to those counter times or locate points to find songs you
want to mix. Also note where any level changes occurred so you can
compensate during mixdown. It helps to note a counter time when
the signal level was very high. When you mix the recording you can
start at that point in setting your overall mix levels.
TEARDOWN
After the gig, pack your mics away first because they may be stolen or
damaged. Refer to your equipment list as you repack everything. Note
equipment failures and fix broken equipment as soon as possible.
437
438
CHAPTER 17 On-Location Recording of Popular Music
After you haul your gear back to the studio, it’s time for mixing and
editing.
1. If necessary, copy all the tracks into your recording software’s
multitrack template.
2. Select all the tracks, and split all the tracks at the same point—
several seconds before and after each song.
3. Export only the tracks in each split section to separate song
folders. That way you can mix each song separately by importing the tracks from each folder into a multitrack template.
Another option: Copy the tracks in each split section, then
paste them into new projects.
Alternatively, use volume envelopes or automation to mix all the
songs, then export one long concert mix. If you want, you can import
that stereo mix and split it into individual songs, saving only the best
ones. Then overlap one song’s ending applause with the beginning
of the next song.
Some of the information in this chapter was derived from two workshops presented at the 79th convention of the Audio Engineering
Society in October 1985. These workshops were titled “On the
Repeal of Murphy’s Law-Interfacing Problem Solving, Planning, and
General Efficiency On-Location,” given by Paul Blakemore, Neil
Muncy, and Skip Pizzi and “Popular Music Recording Techniques,”
given by Paul Blakemore, Dave Moulton, Neil Muncy, Skip Pizzi, and
Curt Wittig.
CHAPTER 18
On-Location Recording of
Classical Music
This chapter explains how to do an on-location stereo recording of a
classical-music ensemble. We cover both equipment and procedures.
EQUIPMENT
■
■
■
■
■
■
■
■
■
■
■
■
■
■
■
■
■
Microphones (low-noise condenser or ribbon type, omni or
directional, free field or boundary, stereo or matched pair.)
Multitrack recorder or stereo recorder.
Low-noise mic preamps or low-noise mini mixer (necessary if
the mic preamps in your recorder are low quality).
Phantom-power supply (necessary if your mixer or mic preamp
lacks phantom power).
Stereo bar.
Mic stands and booms, or fishing line to hang mics.
Shock mount (optional).
Mixer (necessary if you use more than two mics).
MS matrix box (if you’re recording with the Mid-Side
technique).
Headphones and/or speakers.
Power amplifier for speakers (optional) or powered Nearfield
monitors.
Recording medium: hard drive or Flash memory.
Power strip, extension cords.
Notebook and pen.
Talkback mic and powered speaker (optional).
Tool kit.
Fresh batteries.
439
440
CHAPTER 18 On-Location Recording of Classical Music
First on the list are microphones. You’ll need at least two or three
of the same model number or one or two stereo microphones.
Good microphones are essential, for the microphones—and their
placement—determine the sound of your recording.
For classical-music recording, the preferred microphones are condenser or ribbon types with a wide, flat frequency response and very
low self-noise (explained in Chapter 6). A self-noise spec of less than
20 dB equivalent SPL, A-weighted, is recommended.
If you want to do spaced-pair recording, you can use either omnidirectional or directional microphones. Omnis are preferred because
they generally have a deeper low-frequency response. If you want to
do coincident or near-coincident recording for sharper imaging, use
directional microphones (cardioid, supercardioid, hypercardioid, or
bidirectional). Mics with those polar patterns tend to roll off in the
low frequencies, but you can compensate for this with equalization.
You need a power supply for condenser microphones: either an external phantom-power supply, a mixer or mic preamp with phantom
power, or internal batteries. A low-noise stereo mic preamp (or lownoise portable mixer) lets you make recordings free of electronic hiss.
You can mount the microphones on stands or hang them from the
ceiling with nylon fishing line. Make sure that the fishing line’s tensile strength exceeds the weight of the mics. Check legal safety issues
with hanging mics; different rules apply in different places. Stands are
much easier to set up but more visually distracting at live concerts.
Stands are more suitable for recording rehearsals or sessions with no
audience present. Neumann makes tiltable “auditorium hangers” that
suspend mic capsules of the KM 100 system from their cables.
The mic stands should have a tripod folding base and should extend
at least 13 feet high. Some suitable products are telescoping stands,
which are lightweight and compact. Some examples are the Shure
S15A (www.shure.com), Quik Lok A85 (www.quiklok.com), AEA13MDV (www.wesdooley.com), K&M 21411B (various vendors), and
stands at www.micsupply.com. You can use baby booms or stand
extenders to increase the height of regular mic stands.
A useful accessory is a stereo bar (stereo microphone adapter, stereo
mic mount). This device mounts two microphones on a single stand
for coincident or near-coincident stereo recording. A long stereo bar
On-Location Recording of Classical Music CHAPTER 18
can accommodate spaced-pair miking. Some models are at www
.micsupply.com.
Another needed accessory in most cases is a shock mount to prevent
pickup of floor vibrations. Some models are the On-Stage MY410
(various vendors), Sabra Som SSM-1 (www.sabra-som.com), Shure
A55M (www.shure.com), and AKG H85 and H100 (www.akg.com).
In difficult mounting situations, boundary microphones may come
in handy. They can lie flat on the stage floor to pick up small ensembles or can be mounted on the ceiling or on the front edge of a balcony. They also can be attached to clear Plexiglas panels that are
hung or mounted on mic stands. Boundary mics are made by most
microphone companies.
To monitor in the same room as the musicians, you need some closedcup, circumaural (around the ear) headphones or isolating earphones
to block out the sound of the musicians. You want to hear only what is
being recorded. Of course, the headphones should be wide-range and
smooth for accurate monitoring. A better monitoring arrangement
might be to set up powered Nearfield loudspeakers in a separate room.
If you’re in the same room as the musicians, you have to sit far from
them to clearly monitor what you’re recording. To do that, you need
a pair of 50-foot microphone extension cables. Longer extensions are
needed if the mics are hung from the ceiling or if you monitor in a
separate room.
A mixer is necessary when you want to record more than one source;
for example, an orchestra and a choir, an orchestra with spot mics,
or a band and a soloist. You might put a pair of microphones on
the orchestra and another pair on the choir. Connect the insert sends
from the mixer to a multitrack recorder, then mix the tracks to stereo
back in the studio. If your budget doesn’t allow a multitrack recorder
you could mix the mic signals live to a stereo recorder.
For monitoring a mid-side recording, bring an MS matrix box that
converts the MS signals to left–right signals, which you monitor.
SELECTING A VENUE
If possible, plan to record in a venue with good acoustics. It should
have adequate reverberation time for the music being performed
441
442
CHAPTER 18 On-Location Recording of Classical Music
(about 2 seconds for orchestral music). This is very important,
because it can make the difference between an amateur-sounding
recording and a commercial-sounding one. Try to record in an auditorium, concert hall, or house of worship rather than in a band room
or gymnasium. Halls with wooden surfaces and a shoebox shape tend
to sound best.
You may be forced to record in a hall that is too dead: that is, the
reverberation time is too short. In this case, you may want to add
artificial reverberation from a digital reverb unit or plug-in, or cover
the seats with plywood sheets or 4-millimeter polyethylene plastic
sheeting. Strong echoes can be controlled with carpets, RPG diffusers, or drapes.
If the venue is surrounded by noisy traffic, consider recording after
midnight. Turn off noisy air conditioning if possible while recording.
Call the venue manager and ask that the circuit breakers for the stage
outlets be turned on the day of the session. Ask where you can loadin your equipment. Also make sure that the load-in door will be
unlocked when you plan to load in.
SESSION SETUP
Be sure to test all your equipment for correct operation before going
on the job. If you’re using battery-powered devices, install fresh batteries in them just before the concert. Keep your equipment inside
your home or studio until you’re ready to leave. A recorder left outside in a cold car may become sluggish if the lubricant stiffens, and
batteries may lose some voltage.
Make sure your hard drive or flash card has enough capacity for
the session. Table 13.1 shows the storage required for a 1-hour
recording.
Arrive at the venue a few hours early to allow for setup and for fixing problems.
First, power-up the equipment. You can use batteries or an AC extension cord plugged into an outlet near the stage. Make sure this outlet
is live. Gaffer-tape the extension cord lengthwise or cover it with mats
to prevent tripping. Tie your outlet strip’s AC cord to the extension
cord so they can’t separate if someone pulls on the extension cord.
On-Location Recording of Classical Music CHAPTER 18
If this is a session, listen to the ensemble playing on the stage. Is the
sound bad? You might want to move the musicians out onto the
floor of the hall.
Mounting the Mics
Place your microphones in the desired stereo miking arrangement.
As an example, suppose you’re recording an orchestra rehearsal with
two crossed cardioids on a stereo bar (the near-coincident method).
Screw the stereo bar onto a mic stand and mount two cardioid
microphones on the stereo bar. For starters, angle them 110 degrees
apart and space them 7 inches apart horizontally. Aim them downward so that they’ll point at the orchestra when raised. You may want
to mount the microphones in shock mounts or put the stands on
sponges to isolate the mics from floor vibration.
Basically, place two or three mics (or a stereo mic) several feet in
front of the group, raised up high (as in Figure 18.1). The microphone placement controls the acoustic perspective or sense of distance to the ensemble, the balance among instruments, and the
stereo imaging.
As a starting position, place the mic stand behind the conductor’s
podium, about 12 feet from the front-row musicians. Connect mic
MUSICAL ENSEMBLE
A
MICS
MICS
B
MUSICAL ENSEMBLE
13' - 20'
4' - 20'
FIGURE 18.1
Typical microphone placement for on-location recording of a classical music ensemble: (a) top view,
(b) side view.
443
444
CHAPTER 18 On-Location Recording of Classical Music
cables and gaffer-tape them to the top of each mic stand for strain
relief. Raise the microphones about 14 feet off the floor. This prevents overly loud pickup of the front row relative to the back row of
the orchestra. It also reduces audience noise by mounting the mics
farther from the audience. Gaffer-tape the mic cables to the bottom
of the stand to keep it from being pulled over.
Leave some extra turns of mic cable at the base of each stand so you
can re-position the stands. This slack also allows for people accidentally pulling on the cables. Try to route the mic cables where they
won’t be stepped on, or cover them with mats.
Live, broadcast, or filmed concerts require an inconspicuous mic
placement, which may not be sonically ideal. In these cases, or for
permanent installations, you probably want to hang the microphones
from the ceiling rather than using stands. You can position a stereo
bar with three nylon fishing lines spaced apart. Make sure that the
tensile strength of the fishing line exceeds the weight of the mics and
stereo bar. Hang individual mics by their cables and attach two fishing lines to the front of each mic to aim it. Another inconspicuous
mic placement is on mic-stand booms projecting forward of a balcony. For drama or musicals, directional boundary mics can be placed
on the stage floor near the footlights.
If you’re uncertain which mic technique to use, you could set up two
or more mic arrays and record them to multitrack. When you master
the program, choose the best-sounding technique. You might even
mix the signals of two or more mic pairs.
Connections
Now you’re ready to make connections. Here are some connection
methods for just two mics:
■
■
■
If your stereo recorder has high-quality mic preamps and phantom power, plug the mics directly into the recorder mic inputs.
If your recorder lacks phantom power (or the phantom voltage
is lower than the mics require), plug the mics into a phantom
supply, and from there into your recorder mic inputs.
If your recorder lacks high-quality mic preamps, plug the mics
into a low-noise mic preamp or mixer with phantom power.
Connect cables from there into your recorder line inputs.
On-Location Recording of Classical Music CHAPTER 18
Here are some connection methods for more than two mics:
■
■
■
■
Plug the mics into a stage box (described in Chapter 1) and run
the snake back to your mixer. Plug the snake mic connectors
into your mixer mic inputs. If you’re recording to two-track,
plug the mixer stereo outputs into the recorder line inputs. If
you’re recording to multitrack, plug the mixer insert-send jacks
into the recorder line inputs.
If you have a multichannel audio interface with mic preamps,
plug the mics into the interface. Connect the interface USB or
FireWire port to a laptop computer.
If you want to feed your mic signals to several mixers—
recording, broadcast, PA—plug your mic cables into a mic splitter or distribution amp (described in Chapter 17 under the
heading Splitting the Mic Signals). Connect the splitter outputs
to the snakes for each mixer. Supply phantom from one mixer
only, on the direct side of the split. Each split will have a groundlift switch on the splitter. Set it to ground for the mixer supplying
phantom. Set it to lift for the other mixers (or to ground if that
results in the least hum). This prevents hum caused by ground
loops between the different mixers.
If you’re using directional microphones and want to make their
response flat at low frequencies, you can run them through a
mixer with equalization for bass boost. You might prefer to
equalize the recording during mastering instead. Boost around
50 to 100 Hz until the bass sounds natural or until it matches
the bass response of omni condenser mics. You won’t need
this EQ if the microphones have been factory-equalized for flat
response at a distance.
Monitoring
Put on your headphones or listen over loudspeakers in a separate
room. Sit equidistant from the speakers, as far from them as they are
spaced apart. You’ll probably need to use a Nearfield arrangement
(speakers about 3 feet apart and 3 feet from you) to reduce coloration of the speakers’ sound from the room acoustics.
Turn up the recording-level controls and monitor the signal. When
the orchestra starts to tune up, set the recording levels to peak around
⫺20 dBFS so you have a clean signal to monitor. You can set levels
more carefully later on.
445
446
CHAPTER 18 On-Location Recording of Classical Music
MICROPHONE PLACEMENT
Nothing has more effect on the production style of a classical-music
recording than microphone placement. Miking distance, polar patterns, angling, spacing, and spot miking all influence the recorded
sound character.
Miking Distance
The microphones must be placed closer to the musicians than a
good live listening position. If you place the mics out in the audience where the live sound is good, the recording probably will sound
muddy and distant when played over speakers. That is because all the
recorded reverberation is reproduced up-front along a line between
the monitor speakers, along with the direct sound of the orchestra.
Close miking (4 to 20 ft from the front row) compensates for this
effect by increasing the ratio of direct sound to reverberant sound.
The closer the mics are to the orchestra, the closer it sounds in
the recording. If the instruments sound too close, too edgy, or too
detailed, or if the recording lacks hall ambience, the mics are too
close to the ensemble. Move the mic stand a foot or two farther from
the orchestra and listen again.
If the orchestra sounds too distant, muddy, or reverberant, the mics
are too far from the ensemble. Move the mic stand a little closer to
the musicians and listen again.
Eventually you’ll find a sweet spot, where the direct sound of the
orchestra is in a pleasing balance with the ambience of the concert
hall. Then the reproduced orchestra will sound neither too close nor
too far.
Here’s why miking distance affects the perceived closeness (perspective) of the musical ensemble: The level of reverberation is fairly constant throughout a room, but the level of the direct sound from the
ensemble increases as you get closer to it. Close miking picks up a
high ratio of direct-to-reverberant sound; distant miking picks up a
low ratio. The higher the direct-to-reverb ratio, the closer the sound
source is perceived to be.
If the recording venue is “live” because of hard surfaces — brick,
glass, stone—chances are you will need to mike closely. On the other
On-Location Recording of Classical Music CHAPTER 18
hand, if the venue is “dead” because of soft surfaces—carpet, drapes,
stuffed seats—expect to mike farther away.
An alternative to finding the sweet spot is to place a stereo pair close
to the ensemble (for clarity) and another stereo pair distant from
the ensemble (for ambience). According to Delos recording director
John Eargle, the distant pair should be no more than 30 feet from
the main pair, otherwise its signal might simulate an echo. You
record the two pairs to a multitrack recorder and mix them back in
the studio. The advantages of this method are
■
■
■
■
It avoids pickup of bad-sounding early reflections.
Close miking reduces pickup of background noise.
After the recording is finished, you can adjust the direct/reverb
ratio or the perceived distance to the ensemble.
Comb filtering due to phase cancellations between the two
pairs isn’t severe because the delay between them is great and
their levels and spectra are different.
Similarly, Skip Pizzi recommends a “double MS” technique, which
uses a close MS microphone mixed with a distant MS microphone
(as shown in Figure 18.2). One MS microphone is close to the performing ensemble for clarity and sharp imaging, and the other is
out in the hall for ambience and depth. The distant mic could be
replaced by an XY pair for lower cost.
MUSICAL ENSEMBLE
Spot mics
Main pair
30 ft max.
Distant pair for hall ambience
FIGURE 18.2
Double MS technique using a close main pair and a distant pair for ambience. Spot mics also are shown.
447
448
CHAPTER 18 On-Location Recording of Classical Music
If the ensemble is amplified through a sound-reinforcement system,
you might be forced to mike very close to avoid picking up amplified
sound and feedback from the reinforcement speakers. In that case
you’ll need to add high-quality artificial reverberation or convolution reverb.
For broadcast or communications, consider miking the conductor
with a wireless lavalier mic or stand-mounted mic.
Stereo-Spread Control
Concentrate on the stereo spread. If the monitored spread is too narrow, it means that the mics are angled or spaced too close together.
Increase the angle or spacing between mics until localization is accurate. Note: increasing the angle between mics will make the instruments sound farther away; increasing the spacing won’t.
If off-center instruments are heard far-left or far-right, that indicates your mics are angled or spaced too far apart. Move them closer
together until localization is accurate.
If you record with a mid-side microphone, you can change the monitored stereo spread either during the recording or after.
To change the spread during the recording, connect the stereo-mic
outputs to the matrix box and connect the matrix-box L–R output to
the recorder. Use the stereo-spread control (M/S ratio) in the matrix
box to adjust the stereo spread.
To alter the spread after the recording using a matrix box: Record
the mid signal on one track and the side signal on another track.
Monitor the output of the recorder with a matrix box. Back in the
studio, run the mid and side tracks through the matrix box, adjust
the stereo spread as desired, and record the left and right outputs.
To alter the spread after the recording using a DAW:
1. Record the mid mic on track 1; record the side mic on track 2.
2. Copy or clone track 2 to track 3. Be sure the waveforms are
aligned.
3. Pan track 2 hard left; pan track 3 hard right.
4. Reverse the polarity of track 3 or use an “invert polarity”
plug-in.
On-Location Recording of Classical Music CHAPTER 18
5. Group tracks 2 and 3 so their faders move together.
6. To change the stereo spread, vary the levels of tracks 2 and 3
relative to track 1.
If you’re set up before the musicians arrive, check the localization by
recording yourself speaking from various positions in front of the
microphone pair while announcing your position (e.g., “left side,”
“mid-left,”“center”). Play back the recording to judge the localization accuracy provided by your chosen stereo array. Recording this
localization test is an excellent practice.
MONITORING STEREO SPREAD
Full stereo spread on speakers is a spread of images all the way
between speakers, from the left speaker to the right speaker. Full stereo spread on headphones can be described as stereo spread from
ear to ear. The stereo spread heard on headphones may or may not
match the stereo spread heard over speakers, depending on the
microphone technique used.
Due to psychoacoustic phenomena, coincident-pair recordings have
less stereo spread over headphones than over loudspeakers. Take this
into account when monitoring with headphones or use only loudspeakers for monitoring.
If you’re monitoring your recording over headphones or anticipate
headphone listening to the playback, you may want to use nearcoincident miking techniques, which have similar stereo spread on
headphones and loudspeakers.
Ideally, monitor speakers should be set up in a Nearfield arrangement (say, 3 feet from you and 3 feet apart) to reduce the influence of room acoustics and to improve stereo imaging. On the wall
behind the monitors, attach a panel of acoustic foam that extends a
few feet beyond the speaker spacing.
If you want to use large monitor speakers placed farther away from
you, deaden the control-room acoustics with acoustic foam or thick
fiberglass insulation (covered with muslin). Place the acoustic treatment on the walls behind and to the sides of the loudspeakers. This
smoothes the frequency response and sharpens stereo imaging.
449
450
CHAPTER 18 On-Location Recording of Classical Music
You might include a stereo/mono switch in your monitoring system
to check for mono compatibility.
Soloist Pickup and Spot Microphones
Sometimes a soloist plays in front of the orchestra. You have to capture a tasteful balance between the soloist and the ensemble. That
is, the main stereo pair should be placed so that the relative loudness
of the soloist and the accompaniment is musically appropriate. If the
soloist is too loud relative to the orchestra (as heard on headphones
or loudspeakers), raise the mics. If the soloist is too quiet, lower the
mics. You may want to add a spot mic (accent mic) about 3 feet from
the soloist and mix it with the other microphones. Take care that the
soloist appears at the proper depth relative to the orchestra.
If a sound reinforcement system is in use, place the soloist mic 8 to
12 inches away to prevent feedback. Use a foam windscreen on a
vocal mic. To make the soloist mic inconspicuous, you could place
a small-diaphragm mic at about chest height aiming at the mouth,
and use a slender mic stand such the Schoeps Active Tube RC.
Many record companies prefer to use multiple microphones and
multitrack techniques when recording classical music. Such methods
provide extra control of balance and definition and are necessary in
difficult situations. Often you must add spot or accent mics on various instruments or instrumental sections to improve the balance or
enhance clarity (as shown in Figure 18.2). In fact, John Eargle contended that a single stereo pair of mics rarely works well.
As pkautzsch posted on Gearslutz.com:
“Recording is not about the real picture, but about the picture we
want. I like to have it similar to the real picture—that is, choir behind
orchestra, soloists in front, violins left, etc.—but I’d never try to make
a recording that is real. At the best, it will be credible. What spot mics
do in recording for home stereo use is nothing else than what our eyes
and brains do when attending a concert.”
A choir that sings behind the orchestra can be miked separately with
two to four cardioids. You might place the choir in the audience area
facing the orchestra, and mike the choir.
On-Location Recording of Classical Music CHAPTER 18
If the recording mics are also used for sound reinforcement, place
the choir mics about 3 feet out front and 3 feet above the head
height of the back row. Add piano mics and windscreened soloist
mics about 8 inches from their sound sources. Record all the mics to
multitrack, and mix the tracks back in your studio. Add reverb and
EQ as needed.
Pan each spot mic so that its image position coincides with that of
the main microphone pair. Using the mute switches on your mixing console, alternately monitor the main pair and each spot mic to
compare image positions.
When you use spot mics, mix them at a low level relative to the
main pair—just loud enough to add definition but not loud enough
to destroy depth. Operate the spot-mic faders subtly or leave them
untouched. Otherwise the close-miked instruments may seem to
jump forward when the fader is brought up, then fall back in when
the fader is brought down. If you bring up a spot-mic fader for a
solo, drop it only 6 dB when the solo is over, not all the way off.
Often the timbre of the instrument(s) picked up by the spot mic is too
bright. You can fix it with a high-frequency rolloff or by using a mic with
that characteristic. Adding artificial reverb to the spot mic can help too.
To further integrate the sound of the spots with the main pair, you
might want to delay each spot’s signal to coincide with those of the
main pair. That way, the main and spot signals are heard at the same
time. For each spot mic, the formula for the required delay is
T ⫽
D
C
where
T ⫽ delay time in seconds
D ⫽ distance between each spot mic and the main pair in feet
C ⫽ speed of sound, 1130 ft per second
For example, if a spot mic is 20 feet in front of the main pair, the
required delay is 20/1130 or 17.7 msec. Some engineers add even
more delay (10 to 15 msec) to the spot mics to make them less
noticeable. As a rule of thumb, 1 foot corresponds to about 1 millisecond of delay.
451
452
CHAPTER 18 On-Location Recording of Classical Music
A suggested track assignment for multitrack concert recordings is
shown below:
Tracks 1–2: main pair
Tracks 2–3: distant pair
Tracks 4–5: “outriggers” or widely spaced pair
Tracks 6 and up: spot mics and conductor’s mic (for announcing
takes)
Once the microphones are positioned properly, gaffer-tape the mic
stand legs to the floor (or add sandbags) so that the stands can’t be
knocked over.
SETTING LEVELS
Now you’re ready to set recording levels. Ask the conductor to have
the orchestra play the loudest part of the composition, and set
the recording level for the desired meter reading. A typical recording level is ⫺6 dB maximum on a peak-reading meter for a digital
recorder. The level can go up to 0 dB maximum without distortion,
but aiming for –6 dB allows for surprises. Bass-drum and tympani
hits produce the highest peaks.
If you plan to record a concert with no sound check, you have to
set the record-level knobs to a nearly correct position ahead of time.
Do this during a pre-concert trial recording, or just go by experience: set the knobs where you did at previous sessions (assuming
you’re using the same microphones in the same venue). Another
way to preset the recording level is to aim for a peak meter level of
⫺20 dBFS when the orchestra tunes up.
Before going on location, you could play loud orchestral music over
your studio monitors or home stereo, set up your mics and recorder,
and set approximate recording levels. If you need to set the level less
than one-third up to get a 0 dB meter level, you probably need to
insert a pad or set the mic-gain switch to low gain.
RECORDING A CONCERT
Before the concert obtain a printed program of the musical selections. On this program, next to each piece, you will write down the
recorder counter times of the beginning and end of that piece. That
On-Location Recording of Classical Music CHAPTER 18
will help you locate and identify the pieces correctly when you edit
the recording later.
Start recording when the conductor walks out (or sooner). Record
the concert nonstop if your recording medium allows. Document
your mic placement and recording level for future jobs.
EDITING
At this point, the recording is finished and you brought it back to
your studio. The concert recording has long applause after each
piece. Suppose you want to edit the applause shorter and insert a
few seconds of silence between the compositions. Here is a suggested
procedure for editing a stereo recording using a DAW:
1. Connect a USB cable between the computer and recorder. Then
the computer recognizes the recorder as a storage device. Clickdrag the recorder’s WAVE file of the concert to your hard drive.
2. In the audio editing software, start a new project and set up an
audio track.
3. Import the concert file from the hard drive into the audio track.
4. Play the track and locate the first piece. Refer to the counter
times you wrote on your session notes or concert program.
5. Delete the part of the recording before the first piece, but don’t
close up the space yet.
6. Find the applause at the end of the first piece. About 10 seconds into the applause, split the clip or region. Also split the
clip a few seconds before the start of the next piece. Cut out
the audio between the split points, but don’t close up the
space.
7. Label the clip of the composition with its title.
8. Repeat Step 6 for the rest of the pieces.
Now each musical piece is in a separate, labeled clip or region on
your screen. Next you will add fades and adjust the spacing between
the pieces. Please refer to Figure 18.3.
1. Slide the first piece back to time zero, then slide it to the right
10 frames (1/3 second).
2. At the end of each piece, let the applause play for 3 seconds
then fade it out over about 8 seconds. Use a fade that starts
quickly and ends slowly.
453
454
CHAPTER 18 On-Location Recording of Classical Music
APPLAUSE
END OF
FIRST PIECE
FADE-IN
4-SEC. GAP
BEGINNING OF
SECOND PIECE
FADE-OUT
FIGURE 18.3
A DAW screen showing an applause fade-out, a 4-second gap, and a fade-in before the next piece.
3. If there is background noise such as air-conditioning rumble,
insert a fade-in about 2 seconds before the beginning of each
piece (Figure 18.3). Or you may want the track to start right
when the music starts; i.e., with no ambience before it starts.
4. Time-slide the clips to create a 4-second gap between them (or
whatever interval sounds right).
If you don’t want total silence between pieces, don’t fade out the
applause. End each piece’s clip where the applause stops, and leave
about 4 seconds of recorded “room tone” between clips.
To edit a session that has no applause, set the beginning of each
clip just before the first note. Set the end of each clip just after the
reverberant tail fades to silence. Leave about a 4-second silent gap
between clips (or whatever sounds right).
When you’re finished editing, note the start time and duration
of each clip. You will use this information to write a cuesheet that
determines the start IDs when burning a CD. Export the edited program to a 24-bit stereo file.
Sometimes, despite your best intentions and the finest microphones,
the spectral balance of the recording might be poor. The bass might
On-Location Recording of Classical Music CHAPTER 18
be weak, the strings strident, and so on. Room acoustics play a strong
part in this. Fortunately, a recording having a skewed tonal balance
often can be salvaged with EQ. An effective tool for this purpose is
Harmonic Balancer (www.har-bal.com), which shows the spectrum
of the recording and lets you equalize it as needed.
Once the program is equalized (if necessary), import the equalized
file and normalize it so that the highest peak reaches ⫺0.3 dBFS.
Finally, enable dither and export the normalized mix to a 16-bit
stereo file. You will burn a CD from this file. There’s your finished
recording!
455
This page intentionally left blank
CHAPTER 19
Surround Sound:
Techniques and Media
So far we’ve covered techniques that result in a 2-channel stereo
recording. With stereo, you hear all the instruments and reverb in
front of you, in the area between the two loudspeakers. But with surround sound, you hear audio images in every direction. For example,
the musical ensemble could be up front, while the hall ambience
envelops you from the sides and rear, as it does in real life.
Stereo uses two channels feeding two loudspeakers in front of you.
Surround sound uses multiple channels feeding multiple speakers placed around you. A disadvantage of stereo is that you must
sit in a tiny “sweet spot” to hear correct localization. In contrast,
surround sound can be heard correctly in a wide area between the
speakers.
Surround gives a wonderfully spacious effect. It puts you inside the
concert hall with the musicians. You and the music occupy the same
space—you’re part of the performance. For this reason, some engineers feel that surround is more musically involving, more emotionally intense, than regular stereo. There’s a sense of envelopment.
Surround mixing is fast becoming a valuable new tool to offer your
customers. Because many listeners have home theater systems with
multiple speakers, they’re already set up to play surround audio
recordings.
A magazine devoted to multichannel sound production is Surround
Professional found at www.surroundpro.com/articles/publish/article_
190.shtml.
457
458
CHAPTER 19 Surround Sound: Techniques and Media
SURROUND SPEAKER ARRANGEMENT
Inherited from the film industry, surround sound uses six channels feeding six speakers placed around the listener. This forms a 5.1 surround
system, where the “point 1” is the subwoofer (LFE) channel. The LFE
channel is band-limited to 125 Hz and below, while the other channels are full bandwidth (20 Hz to 20 kHz). “5.1” is a channel format—a
surround-sound standard that states the number of channels, their frequency response, and the speaker placement.
The six speakers are:
1. Left-front
2. Center
3. Right-front
4. Left-surround
5. Right-surround
6. Subwoofer
Figure 19.1 shows the recommended placement of monitor speakers for 5.1 surround sound. It is the standard setup proposed by the
International Telecommunication Union (ITU). From the center
FIGURE 19.1
Recommended placement of monitor speakers for 5.1 surround sound.
Surround Sound: Techniques and Media CHAPTER 19
speaker, the left and right speakers should be placed at ⫾30 degrees,
and the surrounds at ⫾110 degrees. (Some engineers prefer 120
to 125 degrees for the surround speakers.) If you’re mixing movie
soundtracks, use dipole speakers for the surrounds, and place them
to the sides (⫾90 degrees).
The left- and right-front speakers provide regular stereo. The surrounds provide a sense of envelopment due to room ambience. They
also allow sound images to appear behind the listener. Deep bass is
filled in by the subwoofer. Because we don’t localize low frequencies
below about 120 Hz, the sub can be placed almost anywhere without
degrading localization.
Originally developed for theaters, the center-channel speaker is
mounted directly in front of the listener. In a home-theater system,
it is placed just above the TV screen, or just below and in front of the
TV screen. This speaker plays center-channel information in mono,
such as dialog.
Why use a center speaker, when two stereo speakers create a phantom
center image? If you use only two speakers and you sit off-center, the
phantom image shifts toward the side on which you’re sitting. But
a center-channel speaker produces a real image, which doesn’t shift
as you move around the listening area. The center speaker keeps the
actors’ dialog on-screen, regardless of where the listener sits.
Also, the phantom center image doesn’t have a flat frequency
response, but a center speaker does. Why is this? Remember that
a center image results when you feed identical signals to both stereo speakers. The right-speaker signal reaches your right ear, but
so does the left-speaker signal after a delay around your head. The
same thing occurs symmetrically at your left ear. Each ear receives
direct and delayed signals, which interfere and cause phase cancellations at 2 kHz and above. A center-channel speaker doesn’t have this
response anomaly.
With a phantom center image, the response is weak at 2 kHz because
of the phase cancellation just mentioned. To compensate, recording
engineers often choose mics with a presence peak in the upper midrange for vocal recording. The center-channel speaker doesn’t need
this compensation.
459
460
CHAPTER 19 Surround Sound: Techniques and Media
For sharpest imaging and continuity of the soundfield, all the speakers should be:
■
■
■
■
■
■
■
The same distance from the listener
The same model (except the sub)
The same polarity
Direct-radiator types
Driven with identical power amps
Matched in sound pressure level with pink noise (the sub is
10 dB louder)
Placed in a room that is left-right symmetrical in the front and
the rear.
If you’re mixing movie soundtracks, the surround speakers should be
dipole designs rather than direct-radiator types. The dipole speakers
project sound forward and backward to create a diffuse effect.
Typically the speakers are 4 to 8 feet from the listener and 4 feet
high. Use a length of string to place the monitors the same distance
from your head. The sub can go along the front wall on the floor. Be
sure that all the speakers sound the same so there is no change in
tonal balance as you pan images around.
A DSP algorithm to reproduce surround sound on headphones has
been developed by Lake Technology. It is licensed to Dolby Labs
under the name Dolby Headphone. This feature is beginning to
appear in consumer receivers and surround processors.
SETTING UP A SURROUND MONITORING SYSTEM
Working in surround, of course, requires more equipment than
working in stereo. You’ll need:
■
■
■
Five “satellite” monitors and a subwoofer, and six channels of
power amplification, or five powered monitors and a powered sub
A sub/satellite crossover (a bass management system); this is
built into some surround monitor packages
A six-channel volume control (hardware or software)
Be sure to include a subwoofer in your monitor system. If you don’t,
you might not hear low-frequency noises that a home listener with a
sub will hear. These noises include breath pops, mic-stand thumps,
air-conditioning rumble, and excessively heavy deep-bass notes.
Surround Sound: Techniques and Media CHAPTER 19
You could use a home-theater surround receiver for power amplification and bass management. It has a single volume control that
simultaneously adjusts the level of all the tracks. Most home-theater
receivers have five amp channels and a line output that feeds a powered subwoofer. The sub’s power amp should be at least 100 watts,
and the receiver should have six analog inputs for your surround mix.
A feature to look for in surround sound receivers is Dolby Digital
and Digital Theater Systems (DTS) decoders (explained later under
the heading Surround Encoding for CD). Receivers labeled “5.1
ready” or “Dolby Digital ready” aren’t Dolby Digital compatible. The
receiver must have Dolby Digital and DTS decoders to play those
formats. However, you don’t need those decoders to do surround
mixes.
Bass Management
In the surround receiver is a “bass management” circuit. Bass management is nothing more than a subwoofer/satellite crossover filter.
It sends the deep lows to the sub, and sends other frequencies to the
five “satellite” speakers that surround you.
A bass management circuit routes frequencies above about 100 Hz
to the five full-range speakers, and routes frequencies below about
100 Hz from all six channels to the subwoofer. In other words, the
bass management circuit routes low-frequency signals from all the
five channels—and the LFE channel—to your sub. By keeping the deep
lows out of the full-range speakers, bass management reduces their
low-frequency distortion and lets them be made relatively small for
home use. Note that bass management affects only what you monitor,
not what you record.
Bass management can be done by a surround-receiver circuit, a standalone box, a special circuit in a subwoofer, or a software plug-in.
You set the bass management crossover frequency as low as possible
to remove the directionality of the bass, and to extend the headroom
of the sub. Typically you’d set the crossover frequency to the frequency where your full-range speakers are down 3 dB on the low end.
If your five main speakers extend down to 20 or 30 Hz, you don’t
need bass management. In some receivers, the bass management
461
462
CHAPTER 19 Surround Sound: Techniques and Media
crossover frequency is adjustable among 120/100/80 Hz—whatever
your system needs for flattest response.
The crossover frequency is 120 Hz with the Dolby Digital standard
and 80 Hz with the DTS standard. However, the Dolby Digital, DTS,
and THX standard crossover frequencies are irrelevant to the frequency of the bass-management filter for the mixing work in your
studio. And the upper frequency limit of the LFE track (125 Hz) has
nothing to do with the crossover frequencies of the bass management
(120 to 40 Hz). As mentioned above, you set the crossover frequency
according to the frequency response of your monitor speakers.
LFE Channel Filtering
The LFE channel lowpass filter used in bass management is a very
steep filter (48 dB per octave), so it causes a lot of phase shift. But
you can turn off the LFE filter in the encoder, and use your own
gentler filter instead. Try 80 Hz, 24 dB per octave. Use any filter you
like, as long as it removes all the energy at 125 Hz. Insert this filter
between your console’s LFE channel output and the input of the
mixdown recorder’s LFE track. In other words, record the LFE track
pre-filtered, and turn off filtering in the encoder.
Surround Mixing Equipment
Here’s what you need to mix in surround:
■
■
■
■
Multitrack recordings in any format: analog tape, digital tape,
hard disk, etc.
If you’re using a mixing console, it needs to have at least six
output channels (also called busses or subgroups). Most digital consoles have a surround matrix, a section set up for mixing
and monitoring in surround.
If you’re using a console, you need an 8-track recorder (MDM
or hard-disk) to record the surround mix. Tracks 1 through
6 record the surround channels, while tracks 7 and 8 record a
separate stereo mix.
If you’re using a DAW for surround mixing, you need an audio
interface with at least 8 outputs. Set up the DAW to feed the 5.1
output channels to six of the interface outputs, then connect
those outputs to your five powered satellites and subwoofer.
This system replaces the surround receiver mentioned earlier.
Surround Sound: Techniques and Media CHAPTER 19
■
If you’re using a DAW, you need surround mixing software.
For example, MOTU’s Digital Performer and Cakewalk’s Sonar
Producer include surround mixing. You could mix to six tracks
on your hard drive, and copy the six resulting WAVE files to a
CD or DVD.
Connections
Now that you know the necessary equipment for surround mixing
(including bass management), you can wire the system together. Two
methods are described below.
Method 1 uses an external multitrack recorder to record the surround mix tracks. Method 2 uses the DAW to record the surround
mix tracks.
Method 1 using an external multitrack recorder: Basically you connect line-level signals from six busses to the associated tracks on your
mixdown recorder. To monitor those six tracks, connect that recorder’s outputs either to a surround receiver, or to a bass-management
filter that feeds six channels of power amps. The receiver or amplifiers drive the five small speakers and the sub.
Figure 19.2 shows the connections. Patch the console’s bus outputs or surround matrix outputs to the inputs of the 8-track mixdown recorder. On the back of this recorder, connect track outputs
1 through 6 to the inputs of the surround receiver or power amp
inputs. On the receiver or power amps, connect the speaker outputs to the speakers. If your subwoofer is self-powered, connect the
Speakers
Console
outputs
8-track
recorder
Bass manager
& vol. control
6-channel
power amp
L
C
R
Ls
Rs
LFE
In Out
In Out
In Out
L
C
R
Ls
Rs
or surround receiver
LFE/
sub
FIGURE 19.2
Mixdown and monitoring system for 5.1 surround with an external multitrack recorder.
463
464
CHAPTER 19 Surround Sound: Techniques and Media
LFE channel line output to the sub line input. If your speakers are
self-powered, connect them to outputs 1 through 6 of your bass
manager.
When you connect your mixer or DAW to the 8-track mixdown recorder,
which signal goes on which track? The most common track assignment
(the Dolby Digital, ITU, and SMPTE standard) is given below:
1.
2.
3.
4.
5.
6.
7.
8.
Left-front
Right-front
Center
LFE or subwoofer channel
Left-surround
Right-surround
Stereo mix left
Stereo mix right
Be sure to label your recording media with the track assignment.
Method 2 using the DAW to record the surround mix tracks:
Connections for powered monitors are shown in Figure 19.3 (top).
COMPUTER
AUDIO
INTERFACE
PCI, USB or
FIREWIRE
POWERED
SATELLITE
MONITORS
CH 1-6
PCI, USB or
FIREWIRE
COMPUTER
POWERED
SUB WITH
BASS MGT.
CROSSOVER
AUDIO
INTERFACE
SURROUND
RECEIVER
PASSIVE
SATELLITE
MONITORS
CH 1-6
PASSIVE OR
ACTIVE SUB
FIGURE 19.3
Mixdown and monitoring system for 5.1 surround, using a DAW as the multitrack recorder. Top: with
powered monitors. Bottom: with passive monitors.
Surround Sound: Techniques and Media CHAPTER 19
Set the overall listening level with your computer surround-monitoring software, and trim individual speaker levels at each speaker.
Connections for passive monitors are shown in Figure 19.3 (bottom). Set the overall listening level with the volume control on the
surround receiver. The bass-management filter is built into the powered sub or into the surround receiver.
Calibration
It’s essential to calibrate your speakers so that their levels match. You
need a sound level meter (a low-cost one from Radio Shack is adequate). Here’s a suggested procedure:
1. Turn on the LFE channel and disable bass management.
2. Set each power amp channel to the same volume setting.
3. Get a pink noise signal from a console noise generator,
receiver, or CD. Connect the signal to a console input. Other
options are a surround calibration DVD, such as the 5.1
Audio Toolkit DVD from www.gold-line.com.
4. Assign that input channel to one surround channel (output
bus or group fader). For example, start with the front-left
channel.
5. Set the console’s or DAW’s bus faders and master faders to
unity gain (the shaded portion of fader travel).
6. Adjust the input fader so the level on the mixer’s meter is
⫺20 dBFS, the standard level for surround mixes. This level
allows for some headroom.
7. Hold the sound level meter at your mixing position and set
it to C-weighting, slow response. Aim the meter’s mic at the
speaker. Set the level of the power-amp channel (or the powered monitor) so you read 85 dBSPL on the meter.
8. Repeat steps 4 through 7 for each channel and its speaker, one
at a time, so that all speakers are putting out the same level.
(In most home-theater receivers is a menu that lets you set the
level of each channel.) If you’re mixing for theater sound with
dipole speakers, set their level to 82 dBSPL instead.
9. Assign pink noise to just the LFE channel so that you hear the
noise coming from the subwoofer. The playback level of the
LFE channel is set 10 dB higher than the main speaker levels
because that’s done in consumer surround systems. Using the
465
466
CHAPTER 19 Surround Sound: Techniques and Media
receiver’s LFE gain control, or the sub’s level control, set the
sub’s SPL to 95 dBSPL if you’re using an RTA, or to 89 dBSPL
if you’re using a sound level meter.
10. After calibration, record your mixes at -6dBFS maximum in
peak meter mode, and set your master monitor level as desired.
Why do the RTA and sound level meter require different SPL settings?
Unlike an RTA, an SPL meter integrates the energy of all the octaves
it is measuring to come up with a single reading. The main speakers
produce 8 to 10 octaves, but the sub produces 2 octaves. The sub’s 2octave energy reads lower on the SPL meter than the main speakers’
8- to 10-octave energy. So, instead of setting the sub’s SPL to 95 dB,
you set it to 89 dB to allow for this energy difference.
It’s important to phase-align the subwoofer’s signal with the fullrange speakers’ signals. Improper alignment can cause a dip in the
system’s frequency response at the crossover frequency. If your sub
has a phase-matching control, send a sine wave to all speakers at
the crossover frequency. Adjust the phase control to get the loudest
sound at the mix position. If the sub has a polarity switch, try it both
ways, and use the position that gives the loudest bass when all the
speakers are playing.
Here’s another calibration method developed by Lorr Kramer of DTS
and surround-sound expert Mike Sokol: Use bandwidth-limited pink
noise (20 to 80 Hz) for the woofer and 500 to 2000 Hz for the main
speakers. Set each speaker to the same SPL, one at a time. Using
500 Hz to 2 kHz pink noise for the main speakers keeps the signal
out of the LFE channel and avoids exciting room modes. And limiting the top end to 2 kHz helps avoid directionality effects of the mic
in the SPL meter.
Now that your monitors are calibrated, do the same for your mixing
console or DAW recording software.
1. Feed pink noise into one channel input module.
2. Route or patch that module’s signal equally to all six surround
output busses.
3. Set each output bus fader to unity gain.
4. At the input module, set the trim to make the output level
⫺20 dB on the console’s meters.
Surround Sound: Techniques and Media CHAPTER 19
5. Trim the console’s ADAT or TDIF outputs (if used) so that all
six tracks read ⫺20 dBFS. (The LFE channel doesn’t get 10 dB
extra gain in the recording path, only in the monitoring path.)
6. On your surround receiver, set the small/large speaker switch
to small, and enable bass management.
RECORDING AND MIXING POP MUSIC FOR
SURROUND
Doing a multitrack recording for later mixdown in surround is
almost identical to recording for mixdown to stereo. Some engineers add a few ambience mics to pick up the studio acoustics. These
ambience tracks are panned to the rear speakers during mixdown.
Many producers like to monitor the original recording session in stereo, not surround, to reduce the technical issues that can slow down
the session. They wait until mixdown to monitor in surround.
In Chapter 12 we showed some ways to mix multitrack recordings to
2-track stereo. Now let’s look at mixing those same recordings to 5.1
surround.
Panning
To place the image of each track in space around you, some consoles
include a pair of pan controls (left/right and front/rear) for each
channel. Others use a trackball, mouse, or joystick. Some digital
audio editing programs, or plug-ins for those programs, permit surround panning as well. Some plug-ins for Pro Tools are Neyrinck’s
Mix 51, Surround Zone, and Audio Research Labs Surround Stage.
Minnetonka’s SurCode is a program to mix 5.1 surround sound,
including a “Build your own mixer” GUI and automated surround
panners (www.minnetonkaaudio.com).
Suppose your console or recording software has 8 output busses, and
you want to pan or move a track between front-left and rear-right.
Assign the track to the front-left and rear-right busses. They should
be odd-even numbered. When you turn the pan knob from left to
right, the sound image should move as desired.
Where should you pan tracks in a surround mix? There are no set
rules. Some producers like to put all the musicians up front, and put
467
468
CHAPTER 19 Surround Sound: Techniques and Media
the hall ambience and applause in the rear. This works especially
well for live concerts and recordings of classical music. This ambience can be created artificially with a digital reverb, or recorded with
mics in the concert hall. Stereo reverb returns are typically sent to
the rear channels. Several multichannel digital reverbs are available,
both as hardware and as software.
Some producers use the rear speakers for background vocals, percussion, horn stabs, or strings, leaving the lead vocal and rhythm instruments in front. A group of five singers could be panned to place
one singer in each speaker, so that the listener is in the middle of
the group. You might put the lead vocal in the center speaker, and
also partly in the surrounds to pull the vocal out toward the listener.
Spreading instruments between left-front and left-surround, and
right-front and right-surround, gives a greater sense of envelopment.
You can place instruments in fixed positions (static panning), or
move them around in space (dynamic panning). Some consoles let
you move a sound by drawing its path in space on your monitor
screen. You might move a sound along a circle, arc, or line. Then save
these pans as part of your automated mix.
Using the Center Speaker
Some engineers prefer to send center tracks equally to the left and right
speakers—not the center speaker. This creates a phantom center image.
However, sending a track to the center speaker gives a better tonal balance and more stable imaging than the center phantom image.
It’s not recommended to send a track only to the center speaker.
Some listeners don’t have a center speaker. So if you send, say, a
vocal to just the center speaker, those listeners without a center
speaker won’t hear it. Also, home listeners can solo the center channel and hear punch-ins on that channel. So many producers route
center tracks mainly to the left and right speakers, and also feed a
little signal to the center speaker (maybe 6 dB down). Some producers don’t use the center speaker at all.
Using the LFE Channel
With music mixes you seldom need to send anything to the LFE subwoofer track—it gets deep bass from the bass management circuit in
Surround Sound: Techniques and Media CHAPTER 19
the listener’s playback system, and at the proper level. Because the
five main channels of 5.1 surround encoders reproduce from 20 Hz
to 20 kHz, there’s no need to put anything musical in the LFE channel. The consumer’s bass management filter will redirect any sub-80Hz energy to their own subwoofer anyway, whether or not you put
this bass signal in the LFE track. On the other hand, suppose the end
listener has a subwoofer/satellite system with six channels, but no
bass management system. In that case, the sound they hear will be
thin in the bass if there is no signal in the LFE channel.
With film mixes, you typically send low-frequency sound effects
(explosions, tornadoes, crashes) to the LFE channel. The LFE track
also helps with DJ mixes and synth mixes that have lots of lows.
Even if you mix only music, you need a subwoofer to hear the very
deep noises (breath pops, room rumble) that a home listener will
hear with a sub.
Downmixing
Downmixing is making a stereo mix from a 5.1 surround mix. It is
done in the consumer’s home theater receiver. In the downmixing
circuit, the left and right surround channels are blended with the left
and right front channels. The center channel is blended equally with
the left and right channels. The LFE channel is either mixed with the
front signals or not used.
Downmixes made this way seldom create a well-balanced stereo mix,
so be sure to check your 5.1 mix for stereo compatibility. Surround
monitoring systems should have a downmix button so you can hear
how your surround mixes will sound when downmixed to stereo by
consumer receivers.
It’s best to do a separate stereo mix and record it on tracks 7 and 8 of
your 8-track mixdown recorder. This stereo mix can be put on DVDAudio discs or Super Audio CDs (explained later in this chapter)
along with the surround mix.
Surround Mix Delivery Format
You will deliver your final 6-track recording to a mastering facility
that will encode it onto a DVD. Supply them either the multitrack
mixdown recording, or a DVD or CD-R with AIFF or WAV files, one
469
470
CHAPTER 19 Surround Sound: Techniques and Media
for each output channel. Be sure to label the recording with the track
contents. Identify each track with a recorded slate (front-left, center,
sub, etc.). Print a 30-second test tone at 1 kHz at ⫺20 dBFS on all
tracks. Also include this information: LFE channel filtered or not, LFE
channel filter frequency, mixdown listening level in dBSPL, sampling
rate, bit resolution, SMPTE format (if included), media formats, program length, intended final audio sampling rate and bit depth, and a
note about any glitches on your master.
SURROUND-SOUND MIC TECHNIQUES
So far we’ve talked about mixing multitrack recordings to surround—
a method intended mainly for pop music. But for classical music,
you can record in surround using five microphones, which capture the spatial character of the concert hall in which the musical
ensemble is playing. Each mic feeds a separate track of a multitrack
recorder.
Surround mic techniques are somewhat different from stereo mic
techniques. In addition to the usual front-left and front-right mics,
you need two surround mics to pick up the hall ambience, and
sometimes a center mic to feed to the center channel. Note that
listening in surround reduces the stereo separation (stage width)
because of the center speaker, but mic techniques for surround are
optimized to counteract this effect.
A number of mic techniques have been developed for recording in
surround. Let’s take a look at them.
Soundfield 5.1 Microphone System
This system is a single, multiple-capsule microphone (Soundfield
ST250 or MKV) and Soundfield Surround Decoder for recording in
surround. The decoder translates the mic’s B-format signals (X, Y, Z,
and W) into L, C, R, LR, RR, and mono subwoofer outputs.
Delos VR2 Surround Miking Method
The late John Eargle, Delos’ director of recording, developed their
VR2 (Virtual Reality Recording) format. Recordings made with this
Surround Sound: Techniques and Media CHAPTER 19
method offer discrete surround. They also are claimed to sound
good in stereo and very good with “steered” analog decoding, such
as Dolby Pro Logic.
In making these recordings, Eargle typically used the mic placement
shown in Figure 19.4. This method employs an ORTF pair in the center, flanked by two spaced omnis typically 12 feet apart. Two house
mics (to pick up hall reverb) are placed about 23 to 52 feet behind
the main pair. These house mics are cardioids aiming at the upper
rear corners of the hall, spaced about 12 feet apart and about 30 feet
high. Spot mics (accent mics) are placed within the orchestra to add
definition to certain instruments.
The mics are assigned to various tracks of a digital multitrack recorder:
■
■
■
■
1 and 2: A mix of the coincident-pair mics, flanking mics, house
mics, and spot mics
3 and 4: Coincident-pair stereo mic
5 and 6: Flanking mics
7 and 8: House mics (surround mics)
FIGURE 19.4
A Delos surround miking method.
471
472
CHAPTER 19 Surround Sound: Techniques and Media
NHK Method
The Japanese NHK Broadcast Center has worked out another surround miking method. They found that, for surround recording, cardioid mics record a more natural amount of reverb than omni mics.
The mics are placed as described below:
■
■
■
■
A center-aiming mic feeds the center channel.
A near-coincident pair feeds front-left and front-right.
Widely spaced flanking mics add expansiveness.
Up to three pairs of ambience mics aim toward the rear.
Figure 19.5 shows the mic placement. The front mics are placed at
the critical distance from the orchestra, where the direct-sound level
matches the reverberant-sound level. Typically, this point is 12 to 15
feet from the front of the musical ensemble and 15 feet above the
floor.
NHK engineers make this recommendation: When you’re monitoring the surround program, the reverb volume in stereo listening
should match the reverb volume in multichannel listening. That is,
when you fold down or collapse the monitoring from 5.1 to stereo,
the direct/reverb ratio should stay the same.
FIGURE 19.5
An NHK surround-sound miking method.
Surround Sound: Techniques and Media CHAPTER 19
The KFM 360 Surround System
Jerry Bruck of Posthorn Recordings developed this elegant surroundmiking method. It is a form of the mid-side (MS) stereo technique.
Bruck starts with a modified Schoeps KFM 6U stereo microphone,
which is a pair of omni mics mounted on opposite sides of a 7-inch
hard sphere. Next to those mics, nearly touching, are two figure-8
mics, one on each side of the sphere, each aiming front and back
(Figure 19.6). This array creates two MS mic arrays aimed sideways
in opposite directions. The mics don’t seriously degrade each other’s
frequency response.
In the left channel, the omni and figure-8 mic signals are summed to
give a front-facing cardioid pattern. They are also differenced to give
a rear-facing cardioid pattern. The same thing happens symmetrically
in the right channel. The sphere, acting as a boundary and a baffle,
“steers” the cardioid patterns off to either side of center, and makes
their polar patterns irregular.
By adjusting the relative levels of the front and back signals, the user
can control the front/back separation. As a result, the mic sounds
like it is moving closer to or farther from the musical ensemble.
According to Bruck:
“The system is revelatory in its ability to recreate a live event. Perhaps most remarkable is the freedom a listener has to move around
and select a favored position, as one might move around in a concert
hall to select a preferred seat. The image remains stable, without a
discernible sweet spot. The reproduction is unobtrusively natural and
convincing in its sense of being there.”
FIGURE 19.6
The KFM 360 Surround Miking System.
473
474
CHAPTER 19 Surround Sound: Techniques and Media
The four mic signals can be recorded on a 4-track recorder for later
matrixing. The figure-8 mics need some EQ to compensate for their
low-frequency rolloff and loss in the extreme highs. To maintain a
good S/N ratio, this EQ can be applied after the signal is digitized.
Five-Channel Microphone Array with Binaural Head
This method was developed by John Klepko of McGill University. It
combines an array of three directional mics with a 2-channel dummy
head (Figure 19.7):
■
■
■
For the front left and right channels: identical supercardioid mics
For the center channel: a cardioid mic
For the surround channels: a dummy head with two pressuretype omni mics fitted into the ear molds
The mics are shock mounted and have equal sensitivity and equal
gains. Supercardioids are used for the front left/right pair to reduce
center-channel buildup. Although the dummy head’s diffraction causes
peaks and dips in the response, it can be equalized to compensate.
FIGURE 19.7
The Klepko surround-sound miking method.
Surround Sound: Techniques and Media CHAPTER 19
During playback, the listener’s head reduces the acoustical crosstalk
that would normally occur between the surround speakers.
According to Klepko:
“The walkaround tests form an image of a complete circle of points
surrounding the listening position. Of particular interest is the imaging
between ⫾30 degrees and ⫾90 degrees. The array produces
continuous, clear images here where other (surround) techniques
fail.”
“The proposed approach is downward compatible to stereo,
although there will be no surround effect. However, stereo headphone
reproduction will resolve a full surround effect due to the included binaural head-related signals. Downsizing to matrix multichannel (5-2-4
in this case) is feasible except that it won’t properly reproduce binaural signals to the rear because of the mono surrounds. As well, some of
the spatial detail recorded by the dummy-head microphone would be
lost due to the usual bandpass filtering scheme (100 Hz to 7 kHz) of
the surround channel in such matrix systems.”
DMP Method
DMP engineer Tom Jung has recorded in surround using a Decca
Tree stereo array for the band and a rear-aiming stereo pair for the
surround ambience (Figure 19.8). Spot mics in the band complete
the miking. The Decca Tree uses three mics spaced a few feet apart,
with the center mic placed slightly closer to the performers. It feeds
the center channel in the 5.1 system.
The rear-aiming mics are either a coincident stereo mic, another
Decca tree, or a spaced pair whose spacing matches that of the Decca
tree outer pair. Jung tries to aim the rear mics at irregular surfaces to
pick up diffuse sound reflections.
Woszcyk Technique (PZM Wedge plus Opposite-Polarity,
180-Degree Coincident-Cardioid Surround Mics)
A recording instructor at McGill University, Wieslaw Woszcyk, developed an effective method for recording in surround that also works
well in stereo. The orchestra is picked up by a PZM wedge made of
two 18 ⫻ 29 inch hard baffle boards angled 45 degrees. A mini omni
mic is mounted on or flush with each board. At least 20 feet behind
475
476
CHAPTER 19 Surround Sound: Techniques and Media
FIGURE 19.8
A DMP surround miking method.
the wedge are the surround mics: two coincident cardioids angled
180 degrees apart, aiming left and right, and in opposite polarity
(Figure 19.9).
According to Woszcyk, his method has several advantages:
■
■
■
Imaging is very sharp and accurate, and spaciousness is excellent due to strong pickup of lateral reflections.
The out-of-phase impression of the surround pair disappears
when a center coherent signal is added.
The system is compatible in surround, stereo, and mono. In
other words, the surround signals don’t phase-interfere with
the front-pair signals. That is because (1) the surround signals
are delayed more than 20 msec, (2) the two mic pairs operate
in separate sound fields, and (3) the surround mics form a bidirectional pattern in mono, with its null aiming at the sound
source.
If a PZM wedge isn’t acceptable because of its size and weight, other
arrays with wide stereo separation may be substituted.
Surround Sound: Techniques and Media CHAPTER 19
FIGURE 19.9
Woszcyk surround miking method.
FIGURE 19.10
Williams’ five cardioid mic array (multichannel microphone array; MMA).
Williams Five Cardioid Mic Array
Michael Williams, an independent audio consultant, worked out
the math to determine the best cardioid microphone arrangement
for realistic reproduction of surround-sound fields. His method is
shown in Figure 19.10.
Double MS Technique
Developed by Curt Wittig and Neil Muncy, the double MS technique
uses a front-facing mid-side mic pair for direct sound pickup and a
477
478
CHAPTER 19 Surround Sound: Techniques and Media
FIGURE 19.11
Double MS technique.
rear MS pair facing away from the front (Figure 19.11). The rear pair
is placed at or just beyond the critical distance of the room where the
reverberant sound level equals the direct sound level. The matrixed
outputs feed front-left, front-right, rear-left, and rear-right speakers.
No center channel mic is specified, but you could use the frontfacing cardioid mic of the front MS pair for this purpose.
Surround Ambience Microphone Array
The surround ambience microphone (SAM) array was developed by
Gunther Theile of the Institute für Rundfunktechnik (IRT). Four cardioid mics are placed 90 degrees to each other and 21 to 25 cm apart.
No center channel is described.
Chris Burmajster Array
Based on extensive listening tests, this array was invented by Chris
Burmajster of Innocent Ear Ltd. It includes an ORTF pair for left-front
and right-front channels, a center mic, and two rear-facing cardioids
angled 90° for the left-rear and right-rear channels. The mics are
mounted on the metal bar shown in Figure 19.12 (not commercially
available). According to the inventor, this arrangement provides solid
central imaging. The rear ambience channels sound coherent with
the front channels, rather than “disjointed” as can happen with mics
Surround Sound: Techniques and Media CHAPTER 19
30 cm (11.8 in)
17 cm (6.7 in)
Center
RF
LF
Hole for
mic stand
30 cm
(11.8 in)
RR
LR
27 cm
(10.6 in)
LF and RF are ORTF. Center mic hangs downward from bar.
LR and RR are rear-facing cardioids angled 90°.
Holes contain mounts for side-addressed mics.
FIGURE 19.12
Mounting bar for Burmajster surround array.
far back in the hall. Details are at http://homepage.ntlworld.com/
chris.burmajster (see Miking for Surround on the Web site).
Ideal Cardioid Arrangement
This system uses a special mic mount with five arms that radiate out
from a center point, like a star. At the end of each arm is a condenser
mic aiming outward from the center. One example includes the
Microtech Gefell INA 5, which uses five M930 mics in shock mounts
(www.microtechgefell.de). In the SPL Atmos 5.1 Surround Recording
System (discontinued), five Brauner condenser mics feed a fivechannel mixing console, which adjusts the mic polar patterns and
offers panning, bass management, and surround monitoring. SPL’s
Web site is www.spl-usa.com. Both systems use the Ideal Cardioid
Arrangement (ICA 5, ITU-775 specification, Figure 19.13) developed
by Volker Henkels and Ulf Herrmann.
The Holophone H2-PRO Surround Mic
This is a surround microphone using several omni mic capsules
flush-mounted in a football-shaped surface. It captures up to eight
479
480
CHAPTER 19 Surround Sound: Techniques and Media
channels of discrete surround sound and has eight XLR connectors.
It can be found on the Web site www.holophone.com.
Sonic Studios DSM-4CS 4-Channel Surround Dummy Head
The Sonic Studios Web site (www.sonicstudios.com) offers a fourmic array that you can put on your head, or on a dummy head, to
record surround.
Slotte Method
Benedict Slotte developed a surround-miking technique intended to
produce very sharp images. The front three mics (Figure 19.14) are
FIGURE 19.13
Ideal Cardioid Arrangement (ICA) used in the Brauner SPL Atmos 5.1 Adjustable Surround Microphone and
the Microtech Gefell INA 5 Surround Microphone.
TO MUSICAL ENSEMBLE
Top view
CENTER
FRONT-LEFT
FRONT-RIGHT
Three supercardioid mics
provide the front signals.
FIGURE 19.14
Slotte method.
Surround Sound: Techniques and Media CHAPTER 19
an optimized near-coincident array using three supercardioid mics.
The array was designed for minimal crosstalk and time differences
between mics. The level and time differences between the left-center
pair are zero for a sound source at the midpoint between them. The
same is true for the right-center pair. In his AES paper, Slotte lists a
range of mic angling, spacing, and relative levels that work well.
Martin Method
Geoff Martin invented a surround technique using two Blumlein
pairs: one for front left-right, and one for surround left-right (Figure
19.15). The goal is high interchannel coherence for direct sounds
and low interchannel coherence for reverberation. Not shown is
the center-channel mic, a figure-eight which is coincident with the
front pair. The center mic can aim straight down to minimize image
distortion, or it can aim forward but located at least 6 cm below the
main pair. Forward aiming is recommended if you want a lot of
direct sound in the center speaker.
MUSICAL ENSEMBLE
FL
90
FR
FRONT PAIR
TOP VIEW
RL
REAR PAIR
FIGURE 19.15
Martin method.
45 - 90
2.3 FEET
OR LESS
RR
481
482
CHAPTER 19 Surround Sound: Techniques and Media
Mike Sokol’s FLuRB Array
This array uses four coincident cardioid mics at 90 degrees to each
other aiming to the front, left, right, and back (Figure 19.16). The
four mic signals feed a matrix processor (Manley FLuRB) that delivers the correct signals for 5.1 surround and up to 8.1 surround. The
array is compact, relatively low cost, and convenient to use. Plus, it
will sum to stereo or mono without phase cancellations.
Stereo Pair Plus Surround Pair
In this method, the center-channel mic is omitted. You use a standard stereo pair of your choice to pick up the musical ensemble, plus
another stereo pair of your choice to pick up the hall ambience. The
hall mics feed the left- and right-surround channels. For example,
two Crown SASS-P MKII microphones can be placed back to back,
separated by several feet.
You might try a hybrid approach for a pop-music concert: Feed the
front speakers a mix of multiple close-up mics on stage, and feed the
rear speakers the signals from a rear-aiming stereo mic.
SOUND
SOURCE
REAR-AIMING
MIC
MATCHED
CARDIOID
MICROPHONES
SIDE RIGHT MIC
SIDE LEFT MIC
FRONT-AIMING MIC
FIGURE 19.16
Mike Sokol’s FLuRB Array.
Surround Sound: Techniques and Media CHAPTER 19
SURROUND MEDIA
So far we’ve discussed surround recordings, and their creation either
by multitrack mixdown or by surround mic techniques. Now let’s
turn our attention to the media that will play those surround recordings. They can be distributed to the public on CD, Super Audio CD
(SACD), DVD, or Blu-ray. Two DVD formats for surround sound are
DVD-Audio and DVD-Video. Let’s examine all of these formats in
detail.
Compact Disc
The compact disc is a 2-channel format that uses linear pulse code
modulation (PCM) encoding. The CD’s bit depth is fixed at 16 bits,
and the sampling rate is fixed at 44.1 kHz.Storage capacity is 650 MB
(74 minutes) or 700 MB (80 minutes). To fit the six channels of surround into the CD’s two channels, two data-reduction encoding
schemes are used: Dolby Digital or DTS. We’ll explain them later
under the heading Surround Encoding for CD.
DVD
Another medium for playing surround recordings is the Digital
Versatile Disc or DVD. It is a high-capacity optical storage medium
the size of a compact disc (4.73 inches diameter). DVD can store
digital data in three formats: audio, video, and computer.
DVD COMPATIBILITY
The DVD player reads:
■
■
■
■
CD
CD-R (in some units)
CD-ROM
DVD-Video disc (video plus audio)
A DVD-Video player can play a DVD-Audio disc if the latter has a
Dolby digital version of the audio in the DVD-Video zone on the disc.
DVD CAPACITY
Compared to a CD, a DVD has much greater capacity due to its
smaller pits and closer tracks. The scanning laser has a shorter
483
484
CHAPTER 19 Surround Sound: Techniques and Media
wavelength than in a CD player, which lets it read DVD’s denser data
stream. In addition, MPEG-2 data compression of the video data
increases the data density on the DVD.
Some DVDs have a single layer of pits; others have a dual layer at different depths. The laser automatically focuses on the required layer.
The Blu-ray Disc records and plays up to 27 GB of data on a singlesided, single-layer CD-sized disc using a 405 nanometer blue-violet
laser. It uses MPEG-2 video-recording format and AC3 and MPEG-1
Layer 2 audio-recording format.
Listed below are the storage capacities of various media:
■
■
■
■
■
■
A compact disc holds up to 700MB.
A single-sided, single-layer DVD holds 4.7 GB (7 times CD
capacity).
A single-sided, dual-layer DVD holds 8.5 GB.
A double-sided, single-layer DVD (DVD-RAM) holds 9.4 GB.
A double-sided, dual-layer DVD holds 17 GB.
A single-sided, single-layer Blu-ray Disc holds up to 27 GB.
A DVD with 4.7-GB capacity is enough to hold a 2-hour, 10-minute
movie plus subtitles.
DVD PLAYERS
Most DVD players have these outputs:
■
■
■
2-channel analog outputs playing a Dolby-Surround-encoded
stereo signal
A digital output that can be connected to an external decoder
for 5.1-channel surround sound (either Dolby Digital or DTS)
6 analog outputs fed from built-in Dolby Digital or DTS decoders (in some units)
DVD DATA RATE
The DVD data transfer rate is 10.1 megabits per second (Mbps).
A 6-channel surround track, with data compression, consumes 384
or 448 kilobits per second (kbps). Linear PCM stereo, at 16 bits and
48-kHz sampling rate, takes 1.5 Mbps. A 96-kHz sample rate doubles the transfer rate to 3 Mps. A 24-bit resolution increases the rate
to 4.5 Mbps.
Surround Sound: Techniques and Media CHAPTER 19
Two forms of DVD are DVD-Video and DVD-Audio, and we’ll look
at them below.
DVD-VIDEO
DVD-Video is a DVD format for videos: movies with a surround-audio
soundtrack. The audio tracks can be Dolby Digital (AC-3) surround; 5.1
channels, with MPEG-1 compressed audio; or they can be 2 channels,
16- to 24-bit, 48- or 96-kHz, linear PCM audio. In other words, you
can have surround sound with compromised fidelity, or 2-channel
sound with excellent fidelity. A DVD disc can be encoded with both
formats, and listeners can choose which one they want to hear.
All players support Dolby Digital (AC-3) surround. Many software
packages for DVD-Video authoring are easy to use and cost under
$70. Some titles are Roxio’s DVDit and MyDVD; and Cyberlink
PowerDVD. Several manufacturers of DVD recorders offer bundled
DVD authoring software.
DVD-AUDIO
DVD-Audio is a DVD format that features audio programs. It also
can have optional still pictures (slide shows), Internet links, visual
interactive menus, on-screen text and lyrics, and about 15 minutes
of video clips.
DVD-Audio discs use PCM encoding. This encoding can be linear,
as on CDs, or “packed” using Meridian Lossless Packing (MLP).
MLP reduces the data rate up to 50%, but without any data loss:
The reproduced signal is identical, bit for bit, with the original
signal. While a standard CD is a 2-channel format fixed at 16-bit/
44.1 kHz resolution, DVD-Audio is a multichannel surround format
with higher resolutions. Compared to a standard CD, DVD-Audio
permits:
■
■
■
Better fidelity due to higher bit depth and higher sampling rate
(up to 24-bit, 96 kHz with six channels, or 176.4 and 192 kHz
with two channels).
Longer playing time in some formats (up to 6 hours of 16-bit,
44.1-K stereo).
More channels (up to six in the 5.1 scheme, allowing surround
sound).
485
486
CHAPTER 19 Surround Sound: Techniques and Media
■
■
■
■
Optional formats in addition to PCM, such as MPEG-2 BC,
DTS, Dolby Digital, or Direct Stream Digital (DSD; the format
used in the Super Audio CD).
Option for different sampling rate and bit depth on different
channels. DVD-Audio allows a wide range of sampling rates,
number of channels, and quantizations. For example, a 4.7GB, single-layer disc can hold a 75-minute program in which
the left-center-right signals are 24-bit/88.2 kHz and the two surround channels are 20-bit/44.1 kHz.
Option for different formats on different tracks. One track
might be 16-bit/96 kHz surround, while another could be
24-bit/192 kHz stereo.
Other optional media formats besides audio. For example,
you can get video content from computer graphics, still photos, tracking over still photos, and mini-DV camera shots of
concerts.
DVD-Audio content is stored in a separate DVD-Audio zone on the
disc (the AUDIO__TS directory). The program can be copy protected
by a digital signature signal and digital watermarking.
Most DVD-Audio players will also play DVD-Videos and CDs, and
some will play Super Audio CDs. A DVD-Video player can play a
DVD-Audio disc if the latter has a Dolby digital version of the audio
in the DVD-Video zone on the disc. The digital outputs on a DVDAudio player include PCM and Dolby Digital. Some units have DTS
and DSD outputs, and all have multichannel analog outputs. Future
players might have FireWire (IEEE 1394) connections.
You compile and record a DVD-Audio disc as you do a CD-R disc:
with software, a hardware disc recorder, and blank discs. Specifically,
you need DVD-Audio authoring software, a DVD-R recorder or DLT
tape backup drive, and blank DVD discs. The write-once discs have
dye on one side like a CD-R. Handling up to 4.7 GB on a single-sided
disc, they can record DVD-Video, DVD-Audio, or DVD-ROM files.
Two major titles of authoring software are Avid Express Studio HD
and Cirlinca DVD Audio solo. Just click the mouse to assemble a
playlist of surround-sound tracks. Another major DVD-authoring
package is discWelder Bronze by Minnetonka software (www
.minnetonkaaudio.com or www.discwelder.com). This low-cost
Surround Sound: Techniques and Media CHAPTER 19
program lets you record PCM stereo files up to 24-bit, 192 kHz and
PCM 5.1 files up to 24-bit, 48 kHz to DVD-Audio, using a computer
DVD recorder.
Blu-ray Disc (BD)
Blu-ray, also known as Blu-ray Disc (BD) is an optical disc format.
It records, rewrites, and plays high-definition video (HD) and stores
huge amounts of data, more than five times that of the DVD, up to
25 GB on a single-layer disc and 50 GB on a dual-layer disc. Blu-ray is
expected to replace DVD, but DVDs will play on Blu-ray players.
Blu-ray supports these audio codecs:
Linear PCM (LPCM)—up to 8 channels of uncompressed audio
(mandatory)
Dolby Digital (DD)—format used for DVDs, 5.1-channel surround sound (mandatory)
Dolby Digital Plus (DD⫹)—extension of Dolby Digital, 7.1channel surround sound (optional)
Dolby TrueHD—lossless encoding of up to 8 channels of audio
(optional)
DTS Digital Surround—format used for DVDs, 5.1-channel surround sound (mandatory)
DTS-HD High Resolution Audio—extension of DTS, 7.1-channel
surround sound (optional)
DTS-HD Master Audio—lossless encoding of up to 8 channels of
audio (optional)
Super Audio CD
An alternative to DVD-Audio is the Super Audio CD developed by
Sony and Philips. It uses the Direct Stream Digital (DSD) process,
which encodes a digital signal in a 1-bit (bitstream) format at a
2.8224 MHz sampling rate. This system offers a frequency response
from DC to 100 kHz with 120 dB dynamic range.
The Super Audio CD has two layers that are read from one side of
the disc. On one layer is a 2-channel stereo DSD program followed
by the same program in six channels for surround sound. On the
other layer is a Red-Book 16-bit/44.1 K program, which makes dual
inventories unnecessary. The 16-bit/44.1 kHz signal is derived from
487
488
CHAPTER 19 Surround Sound: Techniques and Media
Sony’s Super Bit Mapping Direct processing, which downconverts
the bitstream signal with minimal loss of DSD quality. Standard CD
players can play the 16-bit/44.1 kHz layer, and future players would
be able to play the 2- or 6-channel DSD layer for even better sound
quality.
The TASCAM DV-RA1000HD records DSD audio or high-resolution
DVD audio—up to 24-bit/192 kHz—to DVD blanks. This standalone unit also records standard CD-DA, WAV, and DSDIFF files to
CD and DVD discs.
ENCODING SURROUND FOR RELEASE ON
VARIOUS FORMATS
As we said, surround mixes can be released on CD, DVD-Video,
DVD-Audio, or Super Audio CD (SACD). We need to consider how
to get six channels of surround audio onto those formats.
In the past, DVD-Video discs and videocassettes used Dolby
Surround—a matrix encoding system that combined four channels
(left, center, right, surround) into two channels. The Dolby Pro Logic
decoder in the consumer playback system unfolded the two channels
back into four. The surround channel, which was mono and limited
bandwidth, was reproduced over left and right surround speakers.
Unfortunately, matrixing reduced the separation between channels
because it mixed the channels together.
Recent encoding schemes let us encode six channels of surround
audio onto disc, while keeping the channels discrete (not matrixed
together). Let’s look at how this is done for CD, DVD-Video, and
DVD-Audio.
Surround Encoding for CD
A CD has only enough capacity for two channels of audio. So to get
six channels of surround audio onto a CD, data reduction is needed.
Two data-reduction schemes are Dolby Digital and DTS.
Dolby Digital and DTS each use a different lossy encoding process
to reduce the bit rate needed to transmit the six channels via a 2channel bitstream. “Lossy” means that the reproduced signal is missing some data that appeared in the original signal, so there is a slight
loss in sound quality.
Surround Sound: Techniques and Media CHAPTER 19
Why is this data reduction necessary? Audio at 44.1-kHz sampling
frequency and 16-bit word length flows at a rate of about 700 Kbps
for each channel. The rate for six channels is 4.2 Mbps. To handle
this huge data flow, data reduction is needed. The data-reduction
encoder in Dolby Digital is AC-3. It can be used on any number of
channels.
First used in movie theaters, Dolby Digital and DTS are perceptual
coding methods. They use data reduction (data compression) to
remove sounds deemed inaudible due to masking. Dolby Digital’s
AC-3 encoder compresses the data about 12:1, while DTS compresses about 3:1. Both formats offer six discrete channels of digital
surround sound. DTS resolution is 20-bit, while Dolby Digital is 16,
18, or 20 bits (perhaps 24 bits in the future). Dolby’s AC-3 encoder
accepts data at sample rates of 32, 44.1, or 48 kHz; DTS accepts
44.1 kHz. Compared to Dolby Digital, DTS uses less data compression and needs more data storage space and bandwidth. But some
listeners say that DTS sounds more transparent—more like the original discrete tracks that were fed into the DTS encoder. Still, Dolby is
the most common format.
To play a Dolby Digital or DTS recording, you need a newer DVD
player or a CD player with a digital output. Plug the digital output into a decoder, which extracts the six channels of digital audio
and converts them to analog. The decoder has a DSP chip that can
decode both the DTS and Dolby Digital formats. Some surround
receivers have DTS and Dolby Digital decoders built in.
Surround Encoding for DVD-Video
In a DVD-Video disc, much of the disc’s capacity is taken up by video
signals. So this format also can handle only two channels of audio.
DVD-Video uses Dolby Digital’s AC-3 scheme to encode (datareduce) the six channels of surround into two channels. During playback, the two channels are decoded to recover the six channels. As
we said, those six channels are separate or discrete—not matrixed.
Surround Encoding for DVD-Audio
A DVD-Audio disc has lots of capacity for audio signals. The six
channels can be encoded without any data loss by using MLP.
489
490
CHAPTER 19 Surround Sound: Techniques and Media
Table 19.1
Parameters of Surround Media Formats
Format
Sample rate
Resolution
Stereo/surround
systems
CD
44.1kHz
16-bit
Stereo PCM. 16-, 20-, or
24-bit DTS (6 channels)
SACD
2.82MHz
1-bit
DSD (2 or 6 channels)
DVD-V
48kHz
16- or 24-bit
Stereo PCM, Stereo
MPEG-2, Dolby Digital,
or DTS (2 to 6 channels)
DVD-A
48kHz
16-bit
Dolby Digital (6 channels)
96kHz
24-bit
PCM encoded with MLP
(6 channels)
192kHz
24-bit
PCM encoded with MLP
(stereo)
Summary of Media Formats
Let’s summarize the surround encoding schemes used in different
media:
■
■
■
CD uses DTS or Dolby Digital. Both are lossy systems.
DVD-Video uses Dolby Digital, a lossy system.
DVD-Audio uses MLP, a lossless system.
However, a DVD-Audio disc also can include Dolby Digital mixes so
that it will play on CD players and DVD-Video players.
Table 19.1 summarizes the sample rate, the resolution (bit depth),
and the stereo or surround systems associated with various media
formats. Figure 19.17 summarizes the types of data handled by various formats of surround media and players.
Encoding Hardware and Software for CD and
DVD-Video
With the CD and DVD-Video formats, the surround encoding can
be done either by you or by the mastering house that manufactures
the disc. To do it yourself you need a hardware or software encoder.
Some examples of software encoders are Minnetonka Audio’s
SurCode series, Sonic Foundry’s 5.1 Surround Plugin Pack, Sonic
Foundry’s SoftEncode 5.1 Channel, or Universal Audio’s SmartCode
Surround Sound: Techniques and Media CHAPTER 19
SURROUND MEDIA FORMATS
6 digital tracks in
Dolby Digital AC-3 Encoder
(hardware or software)
ENCODERS
6 channels of digital
audio to DVD-A recorder
MEDIA
CD disc
PLAYERS
DTS Encoder
(hardware or software)
DTS
DVD-Video disc
(PCM or MLP)
Dolby Digital
Up to 6 ch.
or DTS
24/96 PCM OR
6 ch. encoded
to 2 ch. with
data reduction
Graphics
DSD
CD player with
digital output
Digital out
2-channel
Dolby Digital
6 ch. encoded OR PCM with no
data reduction
to 2 ch. with
data reduction
6 channels of digital
audio to Blu-ray recorder
Dolby
Digital
(or DTS
in DVD-A
only)
Internal Dolby
Digital decoder
(or DTS
decoder in
DVD-A only)
SACD disc
Blu-ray disc
DSD format
(2.8 MHz, 1-bit)
Linear PCM (LPCM)
Dolby Digital (DD)
Dolby Digital Plus (DD+)
Dolby TrueHD
DTS Digital Surround
DTS-HD High Res. Audio
DTS-HD Master Audio
Digital out
DSD out
External
Dolby Digital
or DTS
decoder
6 ch. of
analog
out
Blu-ray player
SACD player
2 stereo
6 discrete
DSD surround DSD
channels
channels
SACD
2 ch. of
analog
out
OR
2 ch. and 6 ch. DSD on one layer.
2 ch. stereo PCM (16-bit/44.1k)
on 2nd layer.
DVD-A or DVD-V player
PCM
2 channel
16-24 bits
48 or 96 kHz
Dolby Digital / DTS
decoder
Six channels
of analog out
6 channels of digital
audio to SACD recorder
OR
2 encoded surround channels
out to CD or DVD recorder
DVD-Audio disc
16-bit/44.1 kHz stereo PCM
Dolby Digital
(AC-3)
OR
(not for
commercial
release)
OR
2 stereo
PCM
16-bit/44.1k
channels
8 ch. of
analog
out
6 ch. of
analog
out
2 ch. of
analog
out
2 ch. of
analog
out
FIGURE 19.17
Surround media formats.
Pro. If you’re using a hardware encoder, connect its inputs to the outputs of your 8-track mixdown deck, in parallel with the connections
to the monitor system.
Surround-sound expert Mike Sokol suggested a way to put Dolby
Digital or DTS encoded surround mixes on a CD-R. Load the six
audio tracks as WAV files into your computer and feed them to the
encoding software. Select an AC-3 WAV file output or DTS WAV file
output. Encode the tracks. Finally, burn the resulting WAV file onto
a CD-R. To hear the surround mix, connect the CD player’s digital
output to a surround decoder, such as the one in a surround receiver.
Commercially released CDs must use only DTS. Details are in Mike’s
upcoming book titled Surround Sound Production.
You don’t necessarily have to buy an encoder; the encoding can be
done by a mastering house or DVD manufacturer. Simply send them
2 ch. of
analog
out
491
492
CHAPTER 19 Surround Sound: Techniques and Media
your 8-track recording. But you might want to put an encoder in
your monitor system to hear the effect of the encoder as the end listener will hear it.
DVD PRE-MASTERING FORMATS
Here are acceptable 8-track media formats to send to a DVD manufacturer. Put SMPTE on track 8, or put a stereo mix on tracks 7 and 8.
For Dolby Digital, track assignments are 1, left; 2, center; 3, right; 4,
LFE (low-frequency effects or subwoofer); 5, left surround; 6, right
surround; 7, data channel; 8, data channel. For DTS the track assignment is 1, left; 2, right; 3, left surround; 4, right surround; 5, center; 6,
LFE; 7, data; 8, data.
If you want to create the LFE or 5.1 channel, you can mix the five
other channels and feed them through a 120-Hz lowpass filter. Note
that most music-only surround mixes don’t use the LFE channel.
You might have video slide-show data that accompanies the audio.
Store the video on a removable hard drive or send the file via the
Internet. TIFF and bitmap (BMP) formats are acceptable.
DOLBY UNITS FOR DVD MASTERING
This information is for those who produce the actual DVD. DVD
mastering requires MPEG-2 video and Dolby Digital encoders. You
can use either one for multichannel audio on DVD titles. Dolby
Digital soundtracks are compatible with mono, stereo, or Dolby
Pro Logic systems. Dolby encoders and decoders are described on
Dolby’s Web site at www.dolby.com/professional/index.html.
Special thanks to Mike Sokol for his generous technical help with
this chapter.
CHAPTER 20
Web Audio and Online
Collaboration
You’ve recorded a new song and you can’t wait for people to hear
it. You might want to distribute it on the Web, in addition to (or
instead of) CDs.
Why put your music online? Web excerpts of your songs can entice
listeners to buy your CDs. Making songs available on the Web costs
less than duplicating CDs and selling them in stores. Also, distributing online can be done much more quickly than distributing CDs.
Keep in mind, though, that it’s hard to get your songs noticed among
the millions of other titles that are online. Many books and Web
sites offer suggestions on marketing your music.
STREAMING VERSUS DOWNLOADING
There are two basic ways to transmit music on the Internet: streaming files and downloading files. A streaming file plays as soon as
you click on its title. A downloaded file doesn’t play until you copy
the entire file to your hard disk. Streaming audio is heard almost
instantly (after the buffer memory fills), but often sounds muffled
and can be interrupted by net congestion.
The sound quality of streaming audio depends on the speed of the
modem and ranges from funky AM radio (with a 56K modem) to
near-CD quality (with a cable modem). With downloaded audio you
must wait a few seconds to several minutes to download the song.
But when it plays, the sound is high-fidelity and continuous.
493
494
CHAPTER 20 Web Audio and Online Collaboration
DATA COMPRESSION
Audio files that you record on your computer are usually WAV or
AIFF files. They have no data reduction (data compression), so
they take up lots of memory. A 3-minute song recorded at 16-bits,
44.1 kHz consumes about 32 MB. Downloading a 32-MB WAV file
on the Web would take several hours using a 56K modem. So, audio
WAV files intended for Web download are usually data-reduced or
data-compressed by encoding them as MP3, WMA, or other formats.
For example, if you compressed a 3-minute song by 10:1 it would
consume about 3.2 MB and would download 10 times faster than
the equivalent WAV file.
Most types of data compression tend to degrade sound quality. The
sound quality of a compressed-data format depends on its bitrate,
measured in kilobits per second (kbps). The higher the bitrate, the
better the sound, but the greater the file size. A bitrate of 128 kbps
for stereo MP3 files is considered to be a good compromise between
sound quality and file size. At low bitrates (at and below 128 kbps),
you can hear artifacts such as “swirly” cymbals, smeared transients
such as drum hits, a general “phasey” effect, and less treble. At higher
bitrates above 192 kbps, the artifacts start to disappear, and to most
listeners the sound is CD quality.
In short, data compression reduces file size and download time at
the expense of sound quality.
A stereo MP3 file encoded at 128 kbps is 64 kbps per channel. A
mono MP3 file encoded at 64 K is equal in quality to a stereo MP3
file encoded at 128 K. Mono files consume half the file space of stereo
files if both are the same bitrate.
Table 20.1 relates MP3 stereo bitrate to sound quality.
WEB-RELATED AUDIO FILES
You can put audio on the Web in several file formats. Let’s look at
some of the file types in current use.
WAVE (.wav): A standard PC format for audio files. It encodes
sound without any data reduction by using pulse code modulation. WAVE files that are used to make audio CDs are 16-bit,
44.1 kHz. Because WAVE files consume a lot of memory (about
Web Audio and Online Collaboration CHAPTER 20
Table 20.1
Bitrate (CBR)
Data Compression Ratio and Sound Quality of Various
MP3 Bitrates
Compression
Sound quality
64 kbps
20:1
AM radio quality
128 kbps
10:1
Cassette quality
192 kbps
7:1
Near-CD quality
256 kbps
5:1
CD quality
320 kbps
4.4:1
CD quality
32 MB for a 3-minute song in stereo), they are seldom used on
the Internet for songs because they take so long to download.
However, an Internet connection by fiber-optic cable (such as
Verizon’s Fios) is much faster than a standard copper-wire connection. Fios is claimed to offer up to 50 or 30 Mbps downloads and up to 20 or 5 Mbps uploads. With Fios, it’s possible
to transfer WAV files—even hi-res ones—in a short time. For
example, www.itrax.com offers songs for sale online that are
surround files or 24-bit/96 kHz stereo files.
AIFF (.aif or .aiff): Audio Interchange File Format, a standard
Mac format for audio files. Use this format to transfer audio
files between a PC and a Mac. Like WAV files, AIFF files aren’t
data-compressed.
MID (MIDI, or Musical Instrument Digital Interface): This is
a MIDI sequence, not an audio file. A .mid file is a string of
numbers that represent performance gestures, such as note on/
off, which note is played, key velocity, and so on. Chapter 16
describes how to create MIDI files. Because MIDI files don’t
include audio, they are very compact. The user downloads the
MIDI file and imports it into a MIDI sequencer or MIDI/audio
DAW. The sequence generates instrumental music in a synth,
sound module, or soft synth when played. To ensure that the
proper instrument sounds play, most people use a General
MIDI file format.
The following formats use data compression.
MP3 (MPEG Level-1 Layer-3): The most popular format. You can
choose a low bitrate to make a small file with lower sound
495
496
CHAPTER 20 Web Audio and Online Collaboration
quality, or choose a high bitrate to make a larger file with
higher quality.
When encoding an MP3 file, you have a choice of constant bit
rate (CBR) or variable bit rate (VBR). VBR uses more bits on
complex musical passages and fewer bits on simpler passages. CBR encodes every frame at the same bitrate, so it is less
efficient. Use VBR to get a fixed level of quality at the lowest
possible bitrate and file size. CBR is recommended only for
streaming where the bitrate must be fixed. Average bit rate
(ABR) is a compromise between CBR and VBR modes. Use
ABR when you need to know the size of the encoded file but
still allow some variation in the bit rate.
MP3PRO: An improvement over MP3. Songs encoded at 64 kbps
with MP3PRO are said to sound as good as songs encoded
at 128 kbps with MP3. MP3PRO offers faster downloads and
nearly doubles the amount of music you can put on a Flash
memory player. MP3 and MP3PRO files are compatible with
each others’ players, but a free MP3PRO player is needed to
hear MP3PRO’s improvement in sound quality. See www
.mp3prozone.com.
WMA (Windows Media Audio): This popular Microsoft format
provides Digital Rights Management (DSR), which is licensing
technology that provides copy protection. It’s useful for online
music stores. For more information see www.microsoft.com/
windows/windowsmedia. WMA Pro also supports multichannel
and high resolution audio.
Suppose you want to download a 3-minute song. We’ll assume
that the download speed is 90 kbps per second (typical of DSL).
■ The original WAV file of the song is 30.3 MB, CD quality or
better, and downloads in about 5.6 minutes.
■ An MP3 or WMA file at 128 kbps is 3.0 MB, cassette quality,
and downloads in about 33 seconds.
■ An MP3 or WMA file at 320 kbps is 6.9 MB, CD quality, and
downloads in about 1.3 minutes.
RealAudio: This format is used for streaming audio and for
music downloads from RealAudio’s music store. The streaming fidelity depends on modem speed and the current Internet
bandwidth. RealAudio files (.ra or. rm) are often used as short
excerpts or previews of songs.
Web Audio and Online Collaboration CHAPTER 20
OGG (Ogg Vorbis): License-free open software. For a given file
size, Vorbis sounds better than MP3. Vorbis takes up less file
size than MP3 files of equal quality. For more information see
www.xiph.org and www.vorbis.com.
AAC (MPEG Advanced Audio Coding): AAC offers better sound
quality than MP3 at the same bitrate. Many listeners claim that
AAC files made at 128 kbps sound like the original uncompressed audio source. What’s more, AAC supports multichannel audio and a wide range of sample rates and bit depths. It’s
also used with Digital Rights Management technology by helping to control the copying and distributing of music.
Currently, WMA and MP3 are the most popular types. Audio clip 41
at www.elsevierdirect.com/companions/978024081444.
WHAT YOU NEED
To put your music on the Web, you need to download a few pieces
of software, which are low-cost or free:
Ripper: A program that converts audio from a CD or CD-R to a
WAV file. Examples are Exact Audio Copy (www.exactaudiocopy
.de), CDex (http://cdexos.sourceforge.net), and FreeRip3 (www
.mgshareware.com). Not all CD-ROM drives support ripping.
MP3 Encoder: A program that converts a WAV or AIFF file to an
MP3 file. Examples of freeware are RazorLame (www.free-codecs
.com/download/RazorLame.htm) and Lame MP3 Encoder
(www.free-codecs.com/download/Lame_Encoder.htm ).
You need both. Get Lame version 3.97 or later. Another option
is Altomp3 from www.Yuansoft.com. Some recording software
includes an MP3 encoder, and so does Winamp, FreeRip3,
CDex, dBpowerAmp, foobar2000, Audio Mp3 Editor, and Exact
Audio Copy.
ID3 Tag Editor: This program lets you add song title, artist, genre,
and other metadata information to an MP3 file. This information displays on screen when the MP3 file plays. You can add
studio contact information in a comments field. Two freeware
programs are Stamp ID3 Tag Editor and Mp3tag. Several MP3
playback programs include a tag editor.
497
498
CHAPTER 20 Web Audio and Online Collaboration
WMA Encoder: A program that converts a WAV or AIFF file
to a WMA file. It’s a free download from Microsoft at www
.microsoft.com/windows/windowsmedia/forpros/encoder/
default.mspx. Windows Media Technologies has Digital Rights
Management, which limits copies or format conversions.
Ogg Vorbis Encoder (optional) from www.vorbis.com. Converts
CD tracks or WAV files to Vorbis format. For a given file size,
Vorbis sounds better than MP3. Vorbis takes up less file size
than MP3 files of equal quality.
To listen to your music that you put on the Web, you need:
MP3 Player: A program or a device that plays MP3 files. Some
software player examples are Yahoo Music Jukebox from
http://music.yahoo.com/jukebox/, Winamp from www.winamp
.com, Windows Media Player from www.microsoft.com
/windows/windowsmedia, RealPlayer from www.real.com, and
QuickTime in Apple’s OS or www.apple.com/quicktime/.
Portable MP3 players play MP3 files downloaded from the Web.
Windows Media Player: Although intended to play mainly
Windows Media files, this free program plays many file types:
.avi, .asf, .asc, .rmf, .wav, .wma, .wax, .mpg, .mpeg, .m1v, .mp2,
.mp3, .mpa, .mpe, .ra, .mid, .rmi, .qt, .aif, .aifc, .aiff, .mov,
.au, .snd, and .vod. This player is a free download from www
.microsoft.com/windows/windowsmedia/.
A source for players, rippers, and encoders is www.team-mp3.com.
Some digital audio editing programs can output MP3, WMA, and
RealAudio files.
HOW TO UPLOAD COMPRESSED AUDIO FILES
Got everything you need? Let’s go. Basically, here’s what you’ll be
doing:
1. Start with a WAV file of a song on your computer. Or convert
your song from a CD to a WAV file by using a ripper.
2. Edit and process the WAV file to optimize it for the Internet.
3. Then use an encoder to convert the WAV file to an MP3 or
WMA file.
Web Audio and Online Collaboration CHAPTER 20
4. Finally, send (upload) the converted file to a Web site that features music in the MP3 or WMA format.
Let’s go over the procedure in more detail. You can substitute WMA
for MP3 in the procedure below.
1. Start with a WAV file of the song. Edit the start and stop points
of the song. You might want to edit a 30-second excerpt of a
song to use as a preview of your music. If so, add a fade-in and
fade-out to the preview.
2. Next, you might want to process the audio so that it will sound
louder and clearer when played on the Web. To do this, reduce
the bandwidth: apply low cut below 40 Hz and high cut above
15 kHz.You might want to try a higher frequency low cut and
a lower frequency high cut. Apply a little compression or multiband compression, apply peak limiting, then normalize the
song to maximize its level.
Save the edited, processed song as a WAV file (PC) or AIFF file
(Mac) on your hard drive.
3. Use an MP3 encoder to convert the song’s WAV or AIFF file
to an MP3 file on your hard drive. You might want to use a
bitrate of 128 kbps, which many Web sites require. It’s a good
compromise between file size and sound quality.
4. If you want, add the song’s title, artist, and comments using an
ID3 tag editor.
5. Next you will upload your MP3 files to an MP3 server—a
Web site that accepts MP3 files for distribution on the Web.
Examples include: www.mp3.com, www.myspace.com, www
.AWAL.com, and www.digstation.com(for independents only).
The CD duplicator Disc Makers will upload your songs quickly
to digstation.com for a small fee. Also see the MP3 Sites page
at www.team-mp3.com. Some sites offer free downloads of
their music files; others charge the listener so that you make
some income.
iTunes Producer is free software that lets you prepare your
music for submission to iTunes. Use it to encode your music into
an AAC format; enter album, song, and artwork information; and
send your files to Apple for consideration. More information at www
.iTunes.com.
499
500
CHAPTER 20 Web Audio and Online Collaboration
FIGURE 20.1
CD Baby Web site.
As an alternative, you can send your physical CD to CD Baby (www
.cdbaby.com; Figure 20.1), and they will convert it to data-compressed files for you. They are a popular, affordable way for independent artists to sell CDs.
eMusic (www.emusic.com) doesn’t work with unsigned artists, but
they do accept music from independent labels. Usually this is done
through a service such as The Orchard (www.theorchard.com),
INgrooves (www.ingrooves.com), or TuneCore (www.tunecore.com;
Figure 20.2). They can compress and deliver a label’s CD catalog to
all digital music stores. All income from digital sales is combined
and reported to you.
Once you have chosen a site, go there and click on the button
labeled “Artists Only,” “Submit your music,” “Sell your music,” or
something similar. This is where you upload your MP3 files.
Once you sign up and fill out some forms, click on “Upload” (or
whatever is close to that). You can also upload scanned photos of
Web Audio and Online Collaboration CHAPTER 20
FIGURE 20.2
TuneCore Web site.
your band, your album cover, and text describing your band and its
music. Some sites take a few days or weeks to approve your songs.
Congratulations—you’re on the Web!
PUTTING YOUR MUSIC ON YOUR WEB SITE
So far I’ve covered how to upload your songs to an MP3 server. Now
I’ll explain how to put your songs on your own Web site. People visiting your site can click on a song title to hear a streaming song preview, or to download an entire song.
Let’s start by creating a Web page link to each MP3 file. It’s easy. You
can create the link with Web page design software or with one line of
HTML code.
Again, you can substitute WMA for MP3 in the procedure below.
1. Suppose you have a song on CD called “Blues Bash.” Using
ripper software, transfer that song to your hard drive as a WAV
file (PC) or AIFF file (Mac). In this example, we’ll call the file
BLUES.wav. If you have a ripper/encoder, convert that song to
an MP3 or WMA file and go to Step 3.
501
502
CHAPTER 20 Web Audio and Online Collaboration
2. Use an MP3 encoder to convert the WAV file to an MP3 file.
I recommend a setting of 160 kbps bitrate, VBR, stereo. This
setting gives hi-fi sound with a relatively small file, so it’s fairly
quick to download. You now have an MP3 file called BLUES.
mp3.
3. You can add song-title and artist information with an ID3 tag
editor.
4. In your computer, move BLUES.mp3 to the same directory on
your hard drive that your Web page is in.
5. Use Web page design software to open your band’s Web page.
Here we’ll call it PAGE1.HTM.
6. On that page, where you want the song title to appear, type
in the title of the song. In this example it’s “Blues Bash.” Link
that title to BLUES.mp3.
When you left-click on the song title, BLUES.mp3 should load and
play. When you right-click on the song title, you can select “Save
Target As” to copy the MP3 file to a directory on your hard drive.
If you want to use HTML code:
1. Open your Web page with a browser such as Windows Explorer.
2. Select View ⬎ Source. You’ll see the HTML code for that page
in plain text format.
3. Find a spot on the page where you want the song title to
appear. Type in the title of the song and the link to its MP3
file. In HTML it would look something like this:
<a href ⫽ "blues.mp3"> Blues Bash< /a>
For a WMA file it would be
<a href ⫽ "blues.wma">Blues Bash</a>
4. Save and close the text file, go to Windows Explorer, and select
View ⬎ Refresh. You should see the link to the MP3 or WMA
song. When you left-click on the link, you should hear the
song. When you right-click on it, you can select “Save Target
As” to copy the MP3 file to a directory on your hard drive.
Now that your song links are working correctly on your computer,
it’s time to upload them.
Web Audio and Online Collaboration CHAPTER 20
1. Upload (send) PAGE1.HTM and BLUES.MP3 to your Web
server. Make sure that both files go to the same directory. Some
Web servers have an upload page for this purpose. Note: Some
Web servers don’t allow MP3 files. Find the page on the Web
server site that tells what files they permit.
2. Now start your Web browser and go to PAGE1.HTM on your
site. Click “Refresh” in your browser so that you see the PAGE1.
HTM that you just uploaded.
3. Left-click the song title (in this case, “Blues Bash”), and it
should play after the buffer fills. Or right-click the song title
and select “Save Target As.” The song should download over
a few minutes. Then you can play it with your MP3 player or
WMA player in glorious hi-fi stereo.
You might prefer to upload your songs to SNOCAP (www.snocap
.com) (Figure 20.3). They send you code to paste into your Web site
or MySpace. Then viewers of your site can hear a 30-second sample
of a tune and purchase it for download. SNOCAP makes it easy to
sell digital downloads from your own Web site. A similar service is
Micropay at www.monkrat.com.
FIGURE 20.3
The SNOCAP Web site home page (copyright 2008 Snocap, Inc.).
503
504
CHAPTER 20 Web Audio and Online Collaboration
Other good information on marketing your music online is in the
article “Music’s New Messiahs: You!” by Gary Mraz in the April
2008 issue of EQ Magazine. Also see http://newmusicstrategies.com/
ebook/. CD Baby offers marketing advice after you create an account.
FLASH ANIMATION
Adobe Flash—originally called Shockwave Flash or Macromedia
Flash—is a technology to create and display animation or video
with an accompanying sound track in Web sites, games, and mobile
phones. It uses a scripting language called ActionScript. You create
Flash animations with the Adobe Flash Professional multimedia
authoring program, which explains how to import your audio files
into a Flash presentation.
Flash is a subject in itself and is covered in many books. For details,
check out these Web sites: www.adobe.com, www.royaltyfreemusic
.com/tutorials/import-flash-audio.htm, and www.digitalprosound
.com/Htm/Techniques/2000/Oct/Flash.htm.
COLLABORATING BY SHARING FILES
Recording engineers and musicians in different cities can work
together on a common project by sharing files—sending digital
audio files to others in WAV, MP3 or WMA format.
For example, you can send a song mix to a studio musician in
another city, who overdubs their part and sends it back to you as
a WAV file to import back into your multitrack project. Here’s the
procedure:
1. Export a WAV file of the song’s stereo mix.
2. Copy that WAV file to a data CD-R.
3. Mail or FedEx the CD-R to the musician.
Then the musician follows these steps:
1. Import the mix into their recording program.
2. Overdub their part.
3. Export their part as a WAV file.
4. Copy that WAV file to a data CD-R.
Web Audio and Online Collaboration CHAPTER 20
5. Send the CD-R back to the recording engineer.
6. The engineer imports the WAV file on that CD-R into the multitrack project for mixing and editing.
What if one studio uses MacIntosh and another uses PC? CD-Rs and
DVD-Rs made on a Mac computer can’t be read by a PC computer.
Fortunately, we have MAC-PC conversion software, such as TransMac
by acutesystems.com. It lets a PC read a Mac-burned data CD and
extract the WAV files.
You could e-mail MP3 or WMA files instead. If you want to use those
compressed-data formats, encode the MP3 or WMA files at 256 kbps
or higher, VBR, for best sound quality.
Sending WAV, MP3, or WMA files by e-mail can be cumbersome
because some e-mail programs limit attachments to 10 MB. Also, the
person receiving the file attachment will have to wait a long time for
the e-mail to download. A better arrangement is to upload the audio
files to a file-sharing Web site such as www.yousendit.com, www
.dropload.com, www.proaudiobus.com, Digidesign’s Digidelivery, or
Apple’s iChat AV. When the recipient is ready to download the file,
they can go to that Web site and do so.
Sharing Multitrack Projects
Suppose you want to send your multitrack recordings to another studio for mixing. You can send them the hard drive that the recording
was made on, along with instructions for extracting the WAV files.
Or export a WAV file of each track (starting at time zero), and copy
those files to a data CD-R or data DVD-R. Name each file by song
and instrument.
Before you send a multitrack project to another studio, follow these
steps:
1. Make sure each track is named according to its instrument.
2. To reduce file size, delete unused tracks and delete unused
audio files from the session. Some recording software has a
tool called “Consolidate audio files” which puts all the audio
files used in a project in a common folder.
3. Render (bounce or export) each track that has automation,
effects, and edits to a new track that extends from the beginning
505
506
CHAPTER 20 Web Audio and Online Collaboration
4.
5.
6.
7.
8.
9.
to the end of the project. Start each track at time zero and set
its fader to 0 dB before bouncing. Then you will have several
tracks with no edits, and with automation moves and plug-ins
embedded in the audio. Give only those tracks to the other
studio as 24-bit WAV files. If the other studio will do automation or effects, disable those features before bouncing.
Consider using plug-ins common to Windows and MacIntosh.
If the other studio has the same plug-ins you do, also send
them rendered tracks with the plug-ins disabled. Include notes
on the plug-ins you used, along with any presets.
Do screen captures of the mixer view, showing the fader and
pan settings.
Convert all files to the same bit depth.
Bounce any MIDI soft-synth tracks into audio tracks, or export
the MIDI tracks in Standard MIDI File format.
Many DAWs can read OMF (Open Media Format). If you want
to export your session via OMF, first follow the steps above. Then
export your audio mix file using File ⬎ Export ⬎ OMF or something similar. In Pro Tools check the Include-Audio option. Recent
alternatives to OMF are Advanced Authoring Format (AAF), and
Material eXchange Format (MXF). See www.aafassociation.org.
Cakewalk SONAR saves its projects as a .bun (bundle) file,
which includes all the track recordings, edits, balances, and
plug-ins. Another studio with SONAR Producer can import and
reproduce that project except for plug-ins that they don’t have.
For more information, read the “NARAS DAW Guidelines for Music
Production” at http://womb.mixerman.net/showthread.php?t⫽760.
Read the “Pro Tools Guidelines for Music Production” at www
.charlesdye.com/ptguidelines2.0.pdf. Also see www.grammy.com/
Recording_Academy/Producers_And_Engineers/Guidelines/.
A revised standard has been published: AES31-3-2008 “AES standard for network and file transfer of audio—Audio-file transfer
and exchange—Part 3: Simple project interchange.” This standard
provides a way to express audio-edit data in readable text form to
enable simple, accurate computer parsing. Edit precision is sampleaccurate. A TCF presentation method offers readable frame-based
timecodes. All AES standards documents are available at www.aes
.org/publications/standards. Enter “aes31-3” into the Quick Search
window and click “Search.”
Web Audio and Online Collaboration CHAPTER 20
When sending out your mixes for mastering for CD release, a good
format is one stereo WAV file per song recorded at 24 bits/44.1 kHz.
Many studios can handle higher sample rates. Ideally your stereo
mixes are recorded without overall compression or limiting, and
with a recorded level of ⫺6 dBFS to ⫺3 dBFS maximum in peak
meter mode.
FINDING STUDIO MUSICIANS, PRODUCERS, AND
ENGINEERS
At many Web sites you can find online session work, or find online
session musicians to enhance your projects. Using their services, you
can work with people all over the world from the comfort of your
own studio. Some sites accept only musicians who are with major
labels, or they divide their services into major-label musicians and
minor-label or indie musicians.
For example, sessionplayers.com seeks major-label musicians, singers, and engineers to join their group of available studio professionals. In their Player’s Network, they also offer directories of
professional but more affordable session musicians for recording
work and musical collaborations.
Here are some sites where session players and engineers can do online
collaboration (all sites below start with www.): sessionplayers.com
/playersnetwork.htm, ejamming.com, esession.com, musiciansjunction
.com, bandmix.com, talentmatch.com, findmusicians.org, and
shenandoahmusic.com/online-session-musicians.htm.
Also, search Google for these terms: music collaboration technology,
musician finder, musician’s forum, online recording session, and
online session musicians.
507
This page intentionally left blank
Appendix A
dB or Not dB
In the studio, you need to know how to set and measure signal levels
and match equipment levels. You also need to evaluate microphones
by their sensitivity specs. To learn these skills, you must understand
the decibel, the unit of measurement of audio level.
DEFINITIONS
In a recording studio, level originally meant power, and amplitude
referred to voltage. Today, many audio people also define “level” in
terms of voltage or sound pressure, even though this terminology is not
strictly correct. You should know both definitions to communicate.
Audio level is measured in decibels (dB). One dB is the smallest
change in level that most people can hear—the just-noticeable difference. Actually, the just-noticeable-difference varies from 0.1 to
about 5 dB, depending on bandwidth, frequency, program material,
and the individual. But 1 dB is generally accepted as the smallest
change in level that most people can detect. A 6- to 10-dB increase
in level is considered by most listeners to be “twice as loud.” Sound
pressure level, signal level, and change in signal level all are measured in dB. Play audio clip 42 at www.elsevierdirect.com/companions/
9780240811444to hear the effects of various decibel changes on loudness.
Audio clip 43 wraps up the demonstration.
SOUND PRESSURE LEVEL
Sound pressure level (SPL) is the pressure of sound vibration measured at a point. It’s usually measured with a sound level meter in
dB SPL (decibels of sound pressure level).
509
510
APPENDIX A dB or Not dB
FIGURE A.1
A chart of sound pressure levels.
The higher the sound pressure level, the louder the sound (Figure
A.1). The quietest sound you can hear, the threshold of hearing, is
0 dB SPL. Average conversation at 1 foot is 70 dB SPL. Average homestereo listening level is around 85 dB SPL. The threshold of pain—so
loud that the ears hurt and can be damaged—is 125 to 130 dB SPL.
Sound pressure level in decibels is 20 times the logarithm of the
ratio of two sound pressures:
dB SPL ⫽ 20 log P/Pref
where P is the measured sound pressure in dynes/cm2, and Pref is a reference sound pressure: 0.0002 dyne/cm2 (the threshold of hearing).
SIGNAL LEVEL
Signal level also is measured in dB. The level in decibels is 10 times
the logarithm of the ratio of two power levels:
dB ⫽ 10 log P/Pref
where P is the measured power in watts, and Pref is a reference
power in watts.
dB or Not dB APPENDIX A
Recently it’s become common to use the decibel to refer to voltage
ratios as well:
dB ⫽ 20 log V/Vref
where V is the measured voltage, and Vref is a reference voltage.
This expression is mathematically equivalent to the previous one,
because power equals the square of the voltage divided by the circuit
resistance:
dB ⫽ 10 log P1/P2 ⫽ 10 log ((V12 /R)/(V22 /R))
⫽ 10 log (V12 /V22 ) ⫽ 20 log (V1 /V2 )
The resistance R (or impedance) in this equation is assumed to be
the same for both measurements, and thus divides out.
Signal level in decibels can be expressed in various ways, using various units of measurement:
■
■
■
■
dBm: decibels referenced to 1 milliwatt, with no defined load
impedance
dBu: decibels referenced to 0.775 volt
dBv: decibels referenced to 0.7746 volt
dBV: decibels referenced to 1 volt
dBm
If you’re measuring signal power, the decibel unit to use is dBm,
expressed in the equation
dBm ⫽ 10 log P/Pref
where P is the measured power, and Pref is the reference power (1
milliwatt).
For an example of signal power, use this equation to convert 0.01
watt to dBm:
dBm ⫽ 10 log (P/Pref )
⫽ 10 log (0.01/0.001)
⫽ 10
So, 0.01 watt is 10 dBm (10 decibels above 1 milliwatt).
511
512
APPENDIX A dB or Not dB
Now convert 0.001 watt (1 milliwatt) into dBm:
dBm ⫽ 10 log (P/Pref )
⫽ 10 log (0.001/0.001)
⫽0
So, 0 dBm ⫽ 1 milliwatt. This has a bearing on voltage measurement
as well. Any voltage across any resistance that results in 1 milliwatt is
0 dBm. This relationship can be expressed in the equation
0 dBm ⫽ V 2 /R ⫽ 1 milliwatt
where V ⫽ the voltage in volts, and R is the circuit resistance in ohms.
For example, 0.7746 volt across 600 ohms is 0 dBm. One volt across
1000 ohms is 0 dBm. Each results in 1 milliwatt.
Some voltmeters are calibrated in dBm. The meter reading in dBm is
accurate only when you’re measuring across 600 ohms. For an accurate dBm measurement, measure the voltage and circuit resistance,
then calculate:
dBm ⫽ 10 log ((V 2 /R)/0.001)
dBu
Another unit of measurement expressing the relationship of decibels
to voltage is dBu. This means decibels referenced to 0.775 volt.
dBu ⫽ 20 log V/Vref
where Vref is 0.775 volt.
dBv
The unit dBv with a small “v” means decibels referenced to 0.7746
volt. This figure comes from 0 dBm, which equals 0.7746 volt across
600 ohms (because 600 ohms used to be a standard impedance for
audio connections):
dBv ⫽ 20 log V/Vref
where Vref is 0.7746 volt.
dB or Not dB APPENDIX A
dBV
Signal level also is measured in dBV (with a capital V), or decibels
referenced to 1 volt:
dBV ⫽ 20 log V/Vref
where Vref is 1 volt. For example, use this equation to convert 1 millivolt (mV; 0.001 volt) to dBV:
dBV ⫽ 20 log (V/Vref )
⫽ 20 log (0.001/1)
⫽ ⫺60
So, 1 millivolt ⫽ ⫺60 dBV (60 decibels below 1 volt). Now convert 1
volt to dBV:
dBV ⫽ 20 log (1/1)
⫽0
So, 1 volt ⫽ 0 dBV.
To convert dBV to voltage, use the formula
Volts ⫽ 10(dBV/20)
Change in Signal Level
Decibels also are used to measure the change in power or voltage
across a fixed resistance. The formula is
dB ⫽ 10 log (P1 /P2 )
or
dB ⫽ 20 log (V1 /V2 )
where P1 is the new power level, P2 is the old power level, V1 is the
new voltage level, and V2 is the old voltage level.
For example, if the voltage across a resistor is 0.01 volt, and it
changes to 1 volt, the change in dB is
513
514
APPENDIX A dB or Not dB
dB ⫽ 20 log (V1 /V2 )
⫽ 20 log (1/0.01)
⫽ 40 dB
Doubling the power results in an increase of 3 dB; doubling the voltage results in an increase of 6 dB.
THE VU METER, ZERO VU, AND PEAK INDICATORS
A VU meter is a voltmeter of specified transient response, calibrated
in volume units or VU. It shows approximately the relative volume
or loudness of the measured audio signal. VU meters are used on
analog tape recorders, broadcast consoles, some live-sound mixing
consoles, and older recording consoles.
The VU-meter scale is divided into volume units, which aren’t necessarily the same as dB. The volume unit corresponds to the decibel only when measuring a steady sine-wave tone. In other words, a
change of 1 VU is the same as a change of 1 dB only when a steady
tone is applied.
Most recording engineers use 0 VU to define a convenient “zero reference level” on the VU meter. When the meter on your mixer or
recorder reads “0” on a steady tone, your equipment is producing
a certain level at its output. Different types of equipment produce
different levels when the meter reads 0 (Figure A.2). Zero VU corresponds to:
■
■
■
■
⫹8 dBm in older broadcast and telephone equipment
⫹4 dBm in balanced recording equipment
⫺10 dBV in unbalanced recording equipment
⫺20 dBFS on a peak-reading meter in a digital recorder or
recording software
A 0 VU recording level (0 on the record level meter) is the normal
operating level of an analog tape recorder; it produces the desired
recorded flux on tape. A “0 VU recording level” doesn’t mean a “0 VU
signal level.”
dB or Not dB APPENDIX A
FIGURE A.2
VU-meter scale.
With a VU meter, 0 VU corresponds to a recording level 8 dB below
the level that produces 3% third-harmonic distortion on tape at
400 Hz. Distortion at 0 VU typically is below 1%.
The response of a VU meter isn’t fast enough to track rapid transients
accurately. In addition, when a complex waveform is applied to a VU
meter, the meter reads less than the peak voltage of the waveform.
(This means you must allow for undisplayed peaks above 0 VU that
use up headroom.)
In contrast, a peak indicator responds quickly to peak program levels, making it a more accurate indicator of recording levels. One type
of peak indicator is an LED that flashes on peak overloads. Another
is the LED bar graph meter commonly seen on digital recorders.
Yet another is the peak program meter (PPM). It is calibrated in dB,
rather than VU. Unlike the VU meter reading, the PPM reading does
not correlate with perceived volume.
In a digital recorder, the meter is an LED or LCD bar graph meter that
reads up to 0 dBFS (FS; full scale). In a 16-bit digital recorder, 0 dBFS
means all 16 bits are ON (1) at the peak of the waveform and all 16
bits are OFF (0) at the trough of the waveform. In a 24-bit digital
recorder, 0 dBFS means all 24 bits are on or off. The OVER indication
means that the input level exceeded the voltage needed to produce
0 dBFS, and there is some short-duration clipping of the output analog
waveform. Some manufacturers calibrate their meters so that 0 dBFS is
less than 16 bits or 24 bits ON; this allows a little headroom.
515
516
APPENDIX A dB or Not dB
BALANCED VERSUS UNBALANCED EQUIPMENT
LEVELS
Generally, audio equipment with balanced (3-pin) connectors
works at a higher nominal line level than equipment with unbalanced (phono) connectors. There’s nothing inherent in balanced or
unbalanced connections that makes them operate at different levels;
they’re just standardized at different levels.
These are the nominal (normal) input and output levels for the two
types of equipment:
■
■
Balanced: ⫹4 dBu (1.23 volts)
Unbalanced: ⫺10 dBV (0.316 volt)
In other words, when a balanced-output recorder reads 0 VU on its
meter with a steady tone, it is producing 1.23 volts at its output connector. This voltage is called ⫹4 dBu when referenced to 1 milliwatt.
When an unbalanced-output recorder reads 0 on its meter with a
steady tone, it is usually producing 0.316 volt at its output connector. This voltage is called ⫺10 dBV when referenced to 1 volt.
INTERFACING BALANCED AND UNBALANCED
EQUIPMENT
There’s a difference of 11.8 dB between ⫹4 dBu and ⫺10 dBV. To find
this, convert both levels to voltages:
dB ⫽ 20 log (1.23/0.316)
⫽ 11.8
So, ⫹4 dBu is 11.8 dB higher in voltage than ⫺10 dBV (assuming the
resistances are the same).
A cable carrying a nominal ⫹4 dBu signal has a signal-to-noise ratio
(S/N) 11.8 dB better than the same cable carrying a ⫺10 dBV signal.
This is an advantage in environments with strong radio-frequency or
hum fields, such as in a computer. But in most studios with short
cables, the difference is negligible.
dB or Not dB APPENDIX A
Connecting a ⫹4 dBu output to a ⫺10 dBV input might cause distortion if the signal peaks of the ⫹4 equipment exceed the headroom
of the ⫺10 equipment. If this happens, use a pad to attenuate the
level 12 dB (Figure A.3, top). The wiring shown converts from balanced to unbalanced as well as reducing the level 12 dB. If you hear
hum with this connection, add an isolation transformer as shown in
Figure A-3, bottom. All the leads should be twisted.
XLR-MALE OR TRS
TO ACTIVE BALANCED CROSS-COUPLED
DEVICE OUTPUT
2-CONDUCTOR
Tip/2 HI
SHIELDED CABLE
Ring/3
LO
Sleeve/1
Shield
not connected
at this end
SHIELD
TO ACTIVE PSEUDO-BALANCED
DEVICE OUTPUT
Tip/2 HI
1-CONDUCTOR
SHIELDED CABLE
4.7K
SLEEVE TIP
PHONE OR RCA PLUG
1.5K TO UNBALANCED
DEVICE INPUT
(plug shell
shown removed)
1K
SLEEVE TIP
Ring/3
Sleeve/1
TO TRANSFORMER-BALANCED
DEVICE OUTPUT
Tip/2 HI
Ring/3
2-CONDUCTOR
SHIELDED CABLE
LO
Sleeve/1
TO BALANCED
DEVICE OUTPUT
Tip/2
HI
Ring/3
SHIELD
2-CONDUCTOR
SHIELDED CABLE
4.7K
SLEEVE TIP
Shield
1.5K
not connected
at this end
1 : 1 TRANSFORMER
4.7K
LO
Sleeve/1
1K
SHIELD
SHIELD
SHIELD
Shield
not connected
at this end
SLEEVE TIP
1.5K
FIGURE A.3
Top: Wiring balanced out to unbalanced in. The resistors form a 12-dB pad to match the balanced
⫹4 dBu output to the unbalanced ⫺10 dBV input. Bottom: Same, with an isolation transformer added to
reduce hum.
517
518
APPENDIX A dB or Not dB
You don’t always need that pad. Many pieces of equipment have a
⫹4/⫺10 level switch. Set the switch to the nominal level of the connected equipment. If there is no such switch on either device, connect between them a ⫹4/⫺10 converter box such as the Ebtech Line
Level Shifter (www.ebtechaudio.com/lls-2des.html) or use the wiring
shown here.
To connect an unbalanced ⫺10 dBV output to a balanced ⫹4 dBu
input, use the wiring shown in Figure A.4. All the exposed leads should
be twisted.
MICROPHONE SENSITIVITY
Decibels are an important concern in another area: microphone sensitivity. A high-sensitivity mic puts out a stronger signal (higher voltage)
than a low-sensitivity mic when both are exposed to the same sound
pressure level.
A microphone-sensitivity specification tells how much output (in
volts) a microphone produces for a certain input (in SPL). The standard is millivolts per pascal, where one pascal (Pa) is 94 dB SPL.
A typical “open-circuit sensitivity” spec is 5.5 mV/Pa for a condenser
mic and 1.8 mV/Pa for a dynamic mic. “Open-circuit” means that
2-CONDUCTOR
SHIELDED CABLE
HI
TIP SLEEVE
PHONE OR RCA PLUG
TO UNBALANCED
DEVICE OUTPUT
(plug shell
shown removed)
3/Ring
LO
Shield
not connected
at this end
1 : 1 TRANSFORMER
2/Tip
1/Sleeve
SHIELD
XLR-MALE OR TRS
TO BALANCED
DEVICE INPUT
2-CONDUCTOR
SHIELDED CABLE
HI
TIP SLEEVE
OUT
Shield
not connected
at this end
2/Tip
3/Ring
LO
SHIELD
1/Sleeve
IN
FIGURE A.4
Top: Wiring unbalanced out to balanced in. Bottom: Same, with an isolation transformer added to reduce hum.
dB or Not dB APPENDIX A
the mic is unloaded (not connected to a load, or connected to a mic
preamp with a very high input impedance). If the spec is 5.5 mV/Pa,
that means the mic produces 5.5 mV when the SPL at the mic is
94 dB SPL.
If you put a microphone in a 20 dB louder soundfield, it produces
20 dB more signal voltage. For example, if 74 dB SPL in gives 0.18 mV
out (⫺75 dBV), then 94 dB SPL in gives 1.8 mV out (⫺55 dBV);
150 dB SPL in gives 1.1 volt out (⫹1 dBV), which is approximately
line level! That’s why you need so much input padding when you
record a kick drum or other loud source.
519
This page intentionally left blank
Appendix B
Optimizing Your Computer for
Multitrack Recording
Once you’ve chosen some recording software and installed it, you’ll
want to make your computer run as fast and glitch-free as possible.
Speed becomes critical as you use more tracks, more effects, high bit
rates, and high sampling rates.
The data flow of multitrack digital audio places high demands on
computer speed. Recording and playback of digital audio requires
long, continuous periods of streaming audio data. The more tracks
in use, and the higher the bit depth and sampling rate, the faster the
data flow has to be. And the more soft synths and effects plug-ins
that are in use, the greater the load on the CPU.
Clearly, you need a fast computer for multitrack recording. But you
also need to optimize the computer’s settings for best results. I will
cover some ways to speed up the data flow and to reduce the CPU
usage. If you follow these tips, you will have a faster system that handles more plug-ins and more tracks at once. Also, when you play tracks
or burn CDs, clicks and drop-outs in the audio will be less likely.
Most of these tips apply to a PC computer. Mac advice is offered at
the end of this appendix in the section Optimizing MacIntosh for
Multitrack Recording.
Disclaimer: Back up your system and data files first, then proceed
at your own risk. Neither I nor Focal Press will be responsible for
damage to your computer system or files from making these changes.
If you aren’t comfortable doing a particular tweak, don’t do it.
However, I have done all these adjustments with no problems, and
they are reversible.
521
522
APPENDIX B Optimizing Your Computer for Multitrack Recording
SPEEDING UP YOUR HARD DRIVE
The hard drive should have a fast average access time (under 9 msec)
and a high internal sustained transfer rate. Some recommended hard
drives are the Seagate Barracuda, Maxtor Diamond Max, and Western
Digital models.
What transfer rate is needed? The data rate of 24 tracks at 24-bits/
96 kHz is 6.6 MB/sec. That is a practical minimum.
The best current hard drives can transfer about 40 MB/sec continuously with a PCI bus and an Ultra ATA/66 or Ultra ATA/100 interface
(for EIDE drives), or an SCSI interface for SCSI drives. SCSI drives
tend to be faster than IDE, but SCSI costs more, and many IDE drives
are fast enough.
To use ATA66 or higher, your motherboard and BIOS must support
it. You need an 80-pin ribbon cable that is 18 inches maximum.
Plug the blue connector into the motherboard, the black one into
the Master device, and the gray one into the Slave device.
Hard drives with high rotational speed or spindle rate (7200 rpm or
greater) tend to have faster transfer rate. They are recommended for
24-bit multitrack productions. The typical sustained transfer rate of
a 7200 rpm drive is 30 to 40 MB/sec. This provides up to 160 16-bit/
44.1 kHz tracks or 50 24/96 tracks, depending on file fragmentation.
For fastest speed, use one hard drive for applications and Windows
files, and another for audio data. That way the audio drive head does
not waste time looking for system files. Put the Windows drive on
the Primary IDE channel as a master, and put the audio drive on the
Secondary IDE channel as a master. To set a drive to master or slave,
change the jumpers on the back of the drives. A laptop computer can
connect to an external USB or FireWire hard drive that is used just
for audio files.
In your recording software under Options, indicate the drive that
contains the audio data.
You can put the two hard drives on the same IDE cable (both on the
Primary channel) as master and slave. Usually, but not always, they
will run a little slower that way. If you have a second CD-ROM drive,
put it on the secondary channel as a slave.
Optimizing Your Computer for Multitrack Recording APPENDIX B
If you have only one hard drive, use one disk partition for audio files
and another partition for programs. That way you can defragment or
reformat the audio data partition frequently.
It’s a good idea to have a dedicated drive just for musical instrument
samples because they generate high bandwidth streams when you
play them.
Defragment the audio file drive often. Defragging reorganizes files into
contiguous areas on the disk so that each audio file can be accessed
with minimum head movements. Before you defrag, disable programs
in the System Tray. Then double-click My Computer, Right-click your
audio drive, and select Properties ⬎ Tools ⬎ Defragmentation.
When you’re finished using all the files on the audio data drive,
reformat it to zero-out the clusters. This is necessary even if you
deleted all the files.
Your system will run faster if you delete unneeded applications.
Select Start ⬎ Settings ⬎ Control panel ⬎ Add/Remove programs.
You must uninstall a program, rather than deleting its folders, to
notice an improvement.
Install lots of RAM, at least 1 GB for audio, or at least 1.5 GB for soft
synths. How does more RAM help? When your DAW application is
streaming audio, it continuously feeds data into a RAM buffer for
each track. If the RAM buffer is big, it needs to be filled less often,
so the disk-drive heads need to search less often. A larger RAM buffer lets the disk heads work more efficiently by getting bigger continuous chunks of data into RAM. Also, having lots of RAM prevents
frequent access of the swap file (described later). The goal is to have
your programs running in RAM, rather than swapping data to your
hard drive.
Important: Check that Direct Memory Access (DMA) is activated on
hard drives and CD ROMs. DMA mode transfers data directly from
the hard drive to RAM, bypassing the CPU. With DMA, your CPU
overhead (while accessing the drive) will fall from about 50 to 5 percent or less. Here is the procedure:
In Windows XP, select Start ⬎ Settings ⬎ Control Panel ⬎ System ⬎
Hardware ⬎ Device Manager ⬎ IDE ATA/ATAPI Controllers (or hard
disk controller) ⬎ Right-click Primary IDE channel ⬎ Properties ⬎
523
524
APPENDIX B Optimizing Your Computer for Multitrack Recording
Advanced Settings Tab ⬎ Transfer Mode: “DMA if available.” Repeat
for the Secondary IDE channel. Reboot.
Disable write-behind caching on your hard drives. Write-behind
caching waits until the system is idle before writing data to the hard
drive, causing a delay. Here’s how to turn it off: In XP select Start ⬎
Settings ⬎ Control Panel ⬎ System ⬎ Hardware ⬎ Device Manager
⬎ Disk drives ⬎ Double-click on your audio data drive ⬎ Policies ⬎
Uncheck “Enable Write Caching” ⬎ OK.
Disable read-ahead caching on your hard drive. If it is enabled, data
are read out of the cache instead of directly from the hard drive—not
good for streaming audio. Your recording program might have a setting for read-ahead caching.
INCREASING PROCESSING SPEED
A computer has a central processing unit (CPU) that does most of
the calculations needed to run software. For example, in some DAWs
the CPU performs all the signal processing for real-time effects. Get a
computer or motherboard with the fastest CPU that you can afford—
a clock speed of 2 Gigahertz or greater. Consider getting a computer
or motherboard with a dual-core or quad-core CPU, an ⫻ 64 processor, a 64-bit operating system, and 64-bit drivers for your hardware.
The Athlon AMD64 CPU or Intel Core 2 Duo are good choices for
audio work. An ⫻ 64 CPU lets the computer carry more RAM, which
helpful for soft synths.
The following tips reduce the amount of data that the CPU has to
handle:
Don’t enable hyperthreading in BIOS unless your software requires it.
If you hear drop-outs or glitches, maybe too many real-time effects
are running. Select a track that has real-time effects, and bounce
(export) that track with effects to an open track. That way the effects
are recorded or embedded in the bounced track, rather than running
in real time. Then archive (save) the original track and delete it from
the project. That reduces CPU loading.
Archive tracks that are muted, then delete those tracks from the project
to further reduce the load on the processor.
Optimizing Your Computer for Multitrack Recording APPENDIX B
It also helps to convert MIDI soft-synth tracks to audio tracks. Some
DAWS have a “freeze” or “bounce MIDI to track” function for this
purpose. Freezing de-activates the synth plug-in, replacing it with an
audio track of the synth output. That frees up processor power for
the mix.
Limit polyphony in a soft synth if you don’t need it.
Don’t insert the same delay effect (echo, reverb, chorus, flanging) in
several tracks. Instead, set up an aux bus that has the desired effect.
Use aux send controls on tracks to adjust the amount of effect on
each track. This reduces the number of effects processes that are running and reduces CPU loading.
In extreme cases, consider installing a Digital Signal Processing
(DSP) card, such as the TC Works PowerCore. It handles the processing for the plug-ins. The plug-ins access the card rather than the
CPU. You can install multiple cards to expand the system. Pro Tools
includes its own DSP cards.
Before recording, turn off the waveform drawing function of your
sound clips or regions.
Reduce the number of colors: Right-click the Desktop ⬎ Properties ⬎
Settings ⬎ set Colors to 16 bit or 256 colors. A higher setting reduces
audio performance when the level meters redraw in your audio
application.
Reduce video acceleration: Select Start ⬎ Settings ⬎ Control panel ⬎
System ⬎ Performance ⬎ Graphics ⬎ Advanced. Reduce Hardware
Acceleration as much as you can without degrading the display of
your audio program.
In XP, adjust visual effects for best performance: Select Start ⬎ Rightclick My Computer ⬎ Properties ⬎ Advanced ⬎ Performance ⬎
Settings ⬎ Visual Effects Tab ⬎ Adjust for best performance.
Important: Change the Performance mode for the operating system:
Right-click My Computer ⬎ Properties ⬎ Advanced ⬎ Performance
⬎ Settings ⬎ Advanced ⬎ Processor scheduling ⬎ select Background
Services. This setting allocates more processor time to background
activities, such as streaming audio and ASIO drivers.
525
526
APPENDIX B Optimizing Your Computer for Multitrack Recording
PREVENTING INTERRUPTIONS
Unnecessary background programs can rob CPU cycles, interrupt
the audio program, and interrupt the hard-drive head from playing
audio. The following tweaks to your PC can increase the number of
tracks and effects you can play without drop-outs or clicks by disabling various background programs or increasing free memory.
Don’t multitask (run other programs) while the audio is streaming.
Each additional task slows down the system. Press Ctrl-Alt-Del to see
what’s running, highlight anything you don’t need, and select End
Task. Leave Explorer and Systray running. Don’t scroll while recording or playing back.
Important: Disable system sounds such as Windows’ chimes and
fanfares: Select Start ⬎ Settings ⬎ Control Panel ⬎ Sounds and
audio devices ⬎ Sounds ⬎ set to No Sounds.
Turn off autosave except when working just on MIDI files.
Use a PS/2 mouse instead of a USB or wireless mouse, which can
cause audio instability.
Switch off your Anti-Virus program when using your programs. Do
this by right clicking on its icon on the bottom right of your screen,
then click “Disable.” You need Anti Virus for downloading files from
outside sources but not for DAW work.
In the preferences menu of your programs, disable auto update to
prevent interruptions. Also select Start ⬎ Settings ⬎ Control Panel ⬎
System ⬎ Automatic Updates ⬎ Turn off automatic updates.
Don’t map through sound card. Select Start ⬎ Settings ⬎ Control
Panel ⬎ Sounds and Audio Devices ⬎ Hardware Tab ⬎ (highlight
your sound card from the list) ⬎ Properties ⬎ Properties Tap ⬎ Audio
Devices ⬎ (highlight your sound card from the list) ⬎ Properties,
and check the “Do not map through this device” checkbox. The
Microsoft device mapper uses unnecessary resources and can add
latency. It’s better to select the audio interface that you want to use
from within your recording software.
Check interrupts. Select Start ⬎ Programs ⬎ Accessories ⬎ System tools
⬎ System information ⬎ Hardware resources ⬎ IRQs. You will see a
complete list of all the IRQs in use. Make sure that SCSI controllers
Optimizing Your Computer for Multitrack Recording APPENDIX B
don’t share interrupts with anything. Make sure the sound card has
its own IRQ number. Don’t share audio cards/devices IRQs with USB,
SCSI, or graphics cards. Don’t share Network/Modems IRQs with
USB/SCSI/Video. Don’t assign IRQ #9 to anything; that is the cascaded IRQ from #2. To reset IRQs, you might need to remove ALL
cards except for the video card and add one card at a time in different
slots.
Place your audio card in a non-IRQ-shared PCI slot. Look in the PCI
table of your motherboard manual for the highest priority slot that
doesn’t share with another device or another slot. Consider using a
shareware utility called PowerStrip, which lets you adjust PCI latency
timers. Set the latencies high enough to get the needed performance,
and no higher.
If you aren’t using any devices on your USB, serial or parallel ports,
go to the BIOS and disable them. This will reduce the number of
IRQs and may prevent conflicts. Also disable or disconnect other
devices that you aren’t using; this frees up interrupts.
Disable background image: Right-click Desktop ⬎ Properties ⬎
Background ⬎ None.
Disable Screen Saver: Right-click Desktop ⬎ Properties ⬎ Screen
Saver ⬎ None.
Important: Switch off power schemes: Start ⬎ Settings ⬎ Control
Panel ⬎ Power Options ⬎ Set “Turn off hard discs” to “Never.”
Switch off hibernation: Start ⬎ Settings ⬎ Control Panel ⬎ Power
Options ⬎ Hibernate ⬎ Uncheck “Enable hibernate support.”
Remove programs and services that load on startup. Some are necessary, but many aren’t. They consume memory and resources. In XP,
select Start ⬎ Run. Type in MSCONFIG and press OK. Select the
Startup tab and disable everything but the system tray and the speaker
volume control. Keep any programs that control hardware. Select the
Services tab. At the bottom of the window, check “Hide all Microsoft
services.” Then press the “Disable All” button. Press OK then Restart.
When Windows reboots, check “Don’t show this message…” and
click OK.
Some experts say that you should NOT disable Multimedia Class
Scheduler, Plug and Play, Superfetch (in Vista), Task Scheduler,
527
528
APPENDIX B Optimizing Your Computer for Multitrack Recording
Windows Audio, and Windows Driver Foundation. If you use Pro
Tools, re-check “MMERefresh” in the Startup tab and “Digidesign
MME Refresh Service” in the Services tab.
SETTING THE BUFFER SIZE
Audio data streaming from your hard drive goes into a buffer memory, then is read out at a constant clock rate. The buffer must be filled
faster than the clock empties it; otherwise there is nothing to read.
This requires the CPU reading from the buffer to pause its processing while the buffer refills. Then you hear a short silence or audio
drop-out. So if the hard drive is too slow, or the buffer is too small,
this causes buffer underrun and results in drop-outs.
If the hard drive streams data faster than the buffer can be emptied,
this causes buffer overflow or overrun. The extra data from the hard
drive might be lost, causing drop-outs, clicks, or stopped playback.
Or the extra data might overwrite adjacent memory locations, causing a system crash.
A small buffer forces the CPU to fill and empty the buffer much
more often, giving the CPU less time for processing. In other words,
the smaller the buffer, the higher the percentage of CPU usage (the
harder the CPU has to work).
So, if you experience drop-outs, crashes, or high CPU percentage,
increase the buffer size. You might start with a setting of 25 msec and
go from there. Don’t make the buffer bigger than necessary, because
then playback will start with a delay.
Another buffer is the I/O buffer, used for reading and writing data to
the hard drive. In your recording software, set disk I/O buffer size to
256 or 512.
MINIMIZING LATENCY
If you’re using soft synths, you want them to respond as fast as possible to MIDI-controller key presses. In other words, you want to minimize latency (monitoring delay). Latency is directly related to buffer
size. Latency (msec) ⫽ buffer size (samples) ⫻ 1000 divided by the
sample rate. For example, if the buffer size is 256 samples and the
sample rate is 44,100 Hz, the latency is 5.8 msec.
Optimizing Your Computer for Multitrack Recording APPENDIX B
A small buffer setting reduces latency—the application responds faster—
but tends to create drop-outs, or reduces the number of tracks that play
without dropouts. A large buffer setting increases latency but prevents
drop-outs. Which setting is best?
To solve this conflict, use two different buffer settings. In your recording program, go to the Preferences menu. Or open your audio
interface’s control window. Set the buffer/latency small (under 4 msec
if possible) while tracking MIDI synths or while processing the input
signal with effects (Echo Input Monitor setting). During playback
and mixdown, set it large (maybe 25 msec). The bigger buffer will
reduce the load on your CPU and let you use more plug-ins.
If you can’t set the latency under 4 msec without drop-outs or
crashes, freeze or bounce tracks that have effects, or temporarily disable any plug-ins. You might also convert MIDI tracks to audio, then
disable soft synths.
Monitoring with headphones has less perceived latency than monitoring with speakers. That’s because sound takes time to travel from
the speakers to your ears. For example, if your speakers are 6 feet
away, that adds 5.3 msec of perceived delay to the monitored signal.
Drivers affect latency, so download the latest drivers for your audio
interface. However, note that earlier versions of drivers are sometimes faster than later versions. Also check out third-party drivers
such as CEntrance Ideal Driver (www.centrance.com) and asio4all, a
universal ASIO driver for WDM audio. You have to set up your audio
software to use ASIO to notice any change in latency.
As we said before, set processor scheduling to “Background Services.”
If you use ASIO drivers, this tweak ensures low latency without
drop-outs.
The round-trip latency of a MIDI keyboard playing a soft synth is
less than the round-trip latency of audio in to audio out. When you
trigger a soft synth with a MIDI keyboard, the round-trip latency is
the sum of these latencies:
■
■
■
Driver (about 4 to 6 msec depending on the driver)
Output buffer (whatever you set it to)
D/A conversion (about 1 to 1.5 msec)
529
530
APPENDIX B Optimizing Your Computer for Multitrack Recording
With the buffer set to 3 msec, the calculated latency when you play
a MIDI soft synth is about 8 msec. That’s not bad, considering that
9 msec is just noticeable.
OTHER TIPS
■
■
■
■
■
■
■
■
■
■
■
Check out www.tweakXP.com, www.Tweak3D.net, and www
.extremetech.com.
Obedia.com offers 24/7 help for computer recording: tech support, consulting and troubleshooting.
More information on computer recording can be found at www
.pcrecording.com.
If you’re using your computer professionally as part of a commercial recording studio, do NOT have a connection to the
Internet on that computer. Have it be a dedicated digital audio
workstation so that viruses, etc., won’t destroy a customer’s session. Use another computer for the Internet.
Four vendors of audio-optimized computers are www
.rainrecording.com, www.dawbox.com, www.sweetwater.com
(Creation Station), and www.liquiddaw.com.
Microsoft has a patch that fixes audio stuttering with USB
devices in XP. Go to www.microsoft.com and obtain Service
Pack 2.
Go to the manufacturer Web sites of your sound card (or audio
interface) and audio editing software. Check out the support
sections for advice on optimizing your computer and on troubleshooting sound problems.
Keep your sound card(s) away from the AGP graphics card to
prevent hum.
Don’t install much software (it can corrupt the registry).
Disable the computer’s internal sound card when you’re using
your own sound card or audio interface. Click Start ⬎ settings ⬎
control panel ⬎ system ⬎ hardware ⬎ device manager ⬎ sound,
video, and game controllers ⬎ Right-click the internal sound
card ⬎ disable. Optional: While you’re in sound controllers,
remove the legacy audio drivers except the ones you’re using.
On the Web, download the latest drivers for your motherboard,
video card, IDE controller, CD burner, recording software, and
sound card or audio interface. Also download the Windows
updates.
Optimizing Your Computer for Multitrack Recording APPENDIX B
■
■
■
■
■
Consider using an AGP video card rather than a PCI video card
to reduce the load on the PCI bus.
Free up disk space by uninstalling Windows components that
you don’t use. Select Start ⬎ Settings ⬎ Control Panel ⬎ Add or
remove programs ⬎ Add/Remove Windows Components.
For XP, disable Disc Indexing Service on NTFS formatted drives
if you do NOT do frequent searches for filenames. Double-click
My Computer ⬎ Right-click each hard drive ⬎ Properties ⬎
Uncheck “Allow Indexing Service to index this disc for fast file
searching” ⬎ Apply. Choose all files and subfolders within the
drive. This action takes a while as all the files are scanned. If
you get a message that says Access Denied…, press “Ignore All.”
Use nLite before installing Windows XP to get a DAW-optimized installation (www.nliteos.com/nlite.html).
If you have a utility that measures disk thruput (sustained disk
transfer rate) you might want to measure your system’s speed
before and after each hard-drive change mentioned above. That
way you will know which changes were effective. A good utility is
DskBench, available free from www.prorec.com and other sources.
WINDOWS VISTA
Windows Vista offers several benefits for recording: improved stability, more processing in the graphics card and less in the CPU, WaveRT
low-latency driver support, and Multimedia Class Scheduler Service
(MMCSS) to allow the audio engine higher priority to the CPU.
Most of the advice for Windows XP applies to Vista. Here are some
additional Vista tips:
■
■
■
■
■
■
Vista Home Premium is recommended for audio (in 32-bit or
64-bit versions).
Be sure to download the latest Vista drivers for your software,
hardware, and plug-ins.
Go to www.update.microsoft.com for performance-increasing
patches.
If your motherboard allows, disable HPET in BIOS.
Use Vista Lite (www.vlite.net) before installing Vista to get a
DAW-optimized installation.
Don’t change the page file (swap file or virtual memory)
because Vista handles it well.
531
532
APPENDIX B Optimizing Your Computer for Multitrack Recording
■
■
■
■
■
■
■
Disable startup items that you don’t need. Select Start ⬎ Run.
Type Defender and press Enter. Select Tools ⬎ Software Explorer
⬎ All Users. Highlight an item you don’t need and click Disable.
Repeat for other items. Reboot when done.
Turn off User Account Control (if you don’t need this security
feature). Select Start ⬎ Control Panel ⬎ User Accounts and
Family Safety ⬎ User Accounts ⬎ Turn User Account Control
on or off ⬎ Uncheck Use User Account Control (UAC) ⬎ OK ⬎
reboot.
Disable Windows components that you don’t need. Select Start
⬎ Run ⬎ appwiz.cpl. Select Turn Windows features on or off.
Unselect any components that you aren’t using. Click OK.
Turn off Windows Aero. Right-click on the desktop and select
Personalize. Select Windows Color and Appearance. Select
“Open classic appearance properties” for more color options.
In the Color Scheme list box, select Windows Vista Basic or
Windows Standard for a more classic appearance. Click OK
Turn off Windows Sidebar. Locate the Sidebar icon in your
systray. Right-click on the icon and select Exit. Uncheck Start
Sidebar when Windows starts. Click OK.
Check www.extremetech.com for Vista hot fixes.
Select Start ⬎ Settings ⬎ Control Panel. Click “Uninstall a
Program” to launch Vista’s “Uninstall or Change a Program
Window”. In the Tasks pane on the left, click “Turn Windows
Features On or Off”. In the list of features, uncheck any you
don’t need.
OPTIMIZING MACINTOSH FOR MULTITRACK
RECORDING
■
■
■
Many of the PC tips apply to Mac: Use a second, fast hard drive
running at 7200 rpm or higher for audio files. Use a minimum
of non-audio software. Disable energy-saver or sleep software.
Update drivers. Reduce monitor resolution. Disable video acceleration software and cards. Check online user groups for advice.
Check the Memory panel and turn off virtual memory.
Check for enough application memory. Select the program’s
icon and press Command-I. In the info window that appears,
try increasing the amount of allotted memory.
Optimizing Your Computer for Multitrack Recording APPENDIX B
■
■
■
■
■
■
■
■
■
■
■
Minimize extensions. Extensions are small programs that extend
the functionality of the operating system when they are loaded.
But they use up CPU clock cycles and slightly reduce stability. You
might disable all extensions but these: sound card, Appearance,
QuickTime, SystemAV, and video drivers. Consider loading the
CD-ROM driver only when you need it. If the recording application freezes after boot-up, check the application’s documentation
for known extension conflicts.
Upgrade to OS X.
For maximum speed, use computers with PCI Express (PCIe).
Use a Mac Pro, an 8-core system.
Use HFS on the audio drive—not HFS (extended). HFS can
make the cluster sizes too small for efficient streaming of audio.
Buy more RAM: 2 GB or more is recommended.
Delete outtakes and unused files to prevent a “too many files
open” message.
If you can’t play enough tracks, increase the buffer size.
If you hear pops or clicks, check your sound card’s clock settings. When recording an analog source, set it to Internal Clock
or Sync. If your source is digital, set it to external S/PDIF sync
or the format you’re using. Be sure sample rates match. Move
audio and digital cables away from USB cables and power lines.
Try turning off USB hubs, modems, and printers.
If your system crashes frequently, try removing downloaded
software. Delete your application’s preferences file if it has
become corrupt.
If the application stops in the middle of a recording, check to
see whether you have set a maximum recording time in advance.
533
This page intentionally left blank
Appendix C
Impedance
Impedance is one of audio’s more confusing concepts. To clarify this
topic, I’ll present a few questions and answers about impedance.
WHAT IS IMPEDANCE?
Impedance (Z) is the resistance of a circuit to alternating current,
such as an audio signal. Technically, impedance is the total opposition (including resistance and reactance) that a circuit has to passing
alternating current.
A high-impedance circuit tends to have high voltage and low current.
A low-impedance circuit tends to have relatively low voltage and
high current.
I’M CONNECTING TWO AUDIO DEVICES. IS IT
IMPORTANT TO MATCH THEIR IMPEDANCES?
WHAT IF I DON’T?
First some definitions. When you connect two devices, one is the source
and one is the load. The source is the device that puts out a signal. The
load is the device you are feeding the signal into. The source has a certain output impedance, and the load has a certain input impedance.
A few decades ago in the vacuum-tube era, it was important to match
the output impedance of the source to the input impedance of the
load. Usually the source and load impedances were both 600 ohms.
If the source impedance equals the load impedance, this is called
“matching” impedances. It results in maximum power transfer from
the source to the load.
535
536
APPENDIX C Impedance
In contrast, suppose the source is low Z and the load is high Z. If
the load impedance is 10 times or more the source impedance, it is
called a “bridging” impedance. Bridging results in maximum voltage transfer from the source to the load. Today, nearly all devices are
connected bridging—low-Z out to high-Z in—because we want the
most voltage transferred between components.
For best sound quality and highest voltage transfer, the input impedance of the connector you’re plugging into should be at least 7 to 10
times higher than the output impedance of the source (such as an
electric guitar or microphone).
For example, if a mic’s impedance is 200 ohms, the input impedance
of a mic input (that you plug the mic into) should be 7 to 10 times
higher, or 1400 to 2000 ohms. If you look at the input impedance
spec for mixer mic inputs, it’s typically around 1500 ohms.
Similarly, the impedance of an electric-guitar pickup is typically
20 to 40 K ohms (20,000 to 40,000) ohms. So the ideal input impedance of a guitar amp (or direct box input, or instrument input) is at
least 7 to 10 times higher, or at least 280 to 400 K ohms.
If you connect a low-Z source to a high-Z load, there is no distortion
or frequency-response change caused by this connection. But if you
connect a high-Z source to a low-Z load, you might get distortion or
altered response. For example, suppose you connect an electric bass
guitar (a high-Z device) into an XLR-type mic input (a low-Z load).
The low frequencies in the signal will roll off, so the bass will sound
thin. And the highs might roll off, making the sound dull.
We want the bass guitar to be loaded by a high impedance, and we
want the mic input to be fed by a low-impedance signal. A direct
box or impedance-matching adapter does this (Figure C.1). Such an
adapter is available from Radio Shack, part no. 274-017.
FIGURE C.1
High-to-low impedance matching adapter.
Impedance APPENDIX C
The adapter is a tube with a phone jack on one end and a male XLR
connector on the other. Inside the tube is a transformer. Its primary
winding is high Z, wired to the phone jack. The transformer’s secondary winding is low Z, wired to the XLR. You plug the guitar cord
into the phone jack, and plug the XLR into a mic input in a snake or
mixer. Use it with a bass guitar, electric guitar, or synth.
This impedance-matching adapter works, but isn’t ideal. The load it
presents to the bass guitar might be 12 kilohms, which will slightly
load down the high-Z guitar pickup, causing thin bass. An active
direct box solves this problem. In place of a transformer, the active
DI usually has a field effect transistor (FET). The FET has a very high
input impedance that doesn’t load down the bass guitar.
In general, the input impedance of a direct box (or an instrument
jack) is higher than the input impedance of a line input.
WHAT ABOUT MICROPHONE IMPEDANCE?
Recording-quality mics have XLR (3-pin) connectors and are low Z
(150 to 300 ohms). A low-Z mic can be used with hundreds of feet
of cable without picking up hum or losing high frequencies.
I’M CONNECTING A MIC TO A MIXER.
IS IMPEDANCE A CONSIDERATION?
Yes. If your mixer has phone-jack inputs, they are probably high Z. But
most mics are low Z. When you plug a low-Z mic into a high-Z input
you get a weak signal. That’s because a high-Z mic input is designed to
receive a relatively high voltage from a high-Z mic, and so the input is
designed to have low gain. So you don’t get much signal amplification.
If you can’t get enough level when you plug a mic into a phone-jack
input, here’s a solution: Between the mic cable and the input jack,
connect an impedance matching adapter (Figure C.2) such as Radio
Shack 274-016. It steps up the voltage of the mic, giving it a stronger
signal.
The adapter is a tube with a female XLR input and a phone-plug output. Inside the tube is a transformer. Its primary winding is low Z, wired
to the XLR. It secondary winding is high Z, wired to the phone plug.
Connect the mic to the XLR; connect the phone plug to the mixer’s
phone jack. Then the mixer will receive a strong signal from the mic.
537
538
APPENDIX C Impedance
FIGURE C.2
Low-to-high impedance matching adapter.
If you’re using a phantom-powered condenser mic, the connections
are different. First, turn off any phantom power in your recordermixer. Connect your mic to a stand-alone phantom-power supply,
and connect the supply output to the impedance-matching adapter.
If your mixer has XLR inputs, they are low-Z balanced. In this case,
simply connect the mic to the mixer using a mic cable with a female
XLR on the mic end and a male XLR on the mixer end. A low-Z mic
input is typically about 1500 ohms, so it provides a bridging load to
a mic that is 150 to 300 ohms.
SHOULD I CONSIDER IMPEDANCE WHEN I CONNECT
TWO LINE-LEVEL DEVICES?
This is seldom a problem. In most audio devices, the impedance of
the line output is low—about 100 to 1000 ohms. The impedance
of the line input is high—about 10 kilohms to 1 megohm. So every
connection is bridging, and you get maximum voltage transfer. Some
audio devices, such as passive equalizers, require a terminating resistor at the input or output for best performance.
CAN I CONNECT ONE SOURCE TO TWO OR MORE
LOADS?
Usually yes. You can connect several devices in parallel across one
line output. Suppose you connect a mixer output simultaneously to
a recorder input, an amplifier input, and another mixer’s input in
parallel. The combined input impedance of those three loads might
be 4000 ohms, which still presents a bridging load to the mixer’s
100-ohm output impedance.
Mics are a different story. Suppose you’re recording a concert. If you
connect one mic to the PA mixer and also to your recording mixer
Impedance APPENDIX C
with a Y cable, the combined input impedance will be about 700
ohms or less. This can load down some microphones, reducing the
bass in dynamic mics or causing distortion in condenser mics. To
prevent the loading, put a 270 ohm 1% resistor in series with Pin 2
and Pin 3 in each leg of the Y cable.
CAN I CONNECT TWO OR MORE SOURCES TO ONE
INPUT?
Not recommended. The low output impedance of one source will
load down the output of the other source, and vice versa. This can
cause level loss and distortion. The same goes for two mics into one
mic input. It’s better to use a small external mixer.
As another option, put a series resistor in line with each device before
combining them. That prevents each device from loading down the
other. A minimum resistor value might be 470 ohms per source. If the
source is balanced, use one resistor on Pin 2 and one on Pin 3—two
1% resistors per source.
SUMMARY
■
■
■
■
■
■
■
■
■
■
■
Impedance (Z) is the opposition to alternating current, measured in ohms.
Microphones and line outputs are usually low Z.
Electric guitars, acoustic-guitar pickups, instrument inputs, and
line inputs are usually high Z.
XLR mic inputs are low impedance; phone jack mic inputs are
high impedance.
Speakers are usually 4 to 8 ohms.
Equal impedances in parallel result in half the impedance.
Equal impedances in series result in twice the impedance.
Connect low-Z sources to low-Z inputs. (A low-Z input is usually 7 to 10 times the source impedance, but it’s still called a
low-Z input.)
Connect high-Z sources to high-Z inputs.
Connect a low-Z source to a high-Z input through a step-up
transformer (impedance matching adapter).
Connect a high-Z source to a low-Z input through a step-down
transformer (impedance matching adapter, or direct box).
539
This page intentionally left blank
Appendix D
Where to Learn More
We recommend the following books, magazines, videos, and literature to anyone who wants more education in recording technology.
BOOKS AND VIDEOS
The Library
Go to your local library and search their database for books on
recording, home recording, music recording, audio recording, MIDI,
or sound production. Read everything you can on the subject.
541
Pro Audio Books Plus
This company offers an online catalog of all sorts of pro audio and
recording books, videos, and courses at www.proaudiobooksplus.com.
Music Books Plus
This is a catalog with a huge number of audio and recording books.
The Web site is www.musicbooksplus.com. They also carry tutorial
videos on DVD.
howaudio.com
This site offers over 1000 tutorial videos on home recording and
related topics.
Focal Press
This is another major audio/recording books catalog and it is found
at http://focalpress.com. Focal Press also offers video DVDs about
542
APPENDIX D Where to Learn More
DAW operation, recording, mic techniques, mixing, and setting up
a home studio. If you want to record concerts or gigs, check out the
Focal Press book Recording Music On Location.
Amazon.com
This online bookstore has a great search engine. In the search field,
type in whatever audio subject you are interested in: DVD, MIDI,
digital audio, recording, mixing, microphones, etc.
RECORDING MAGAZINES
Recording (home and project studio recording), www.recordingmag
.com.
Mix (pro recording and concert sound), www.mixonline.com.
EQ (home, project, and pro recording), www.eqmag.com; EQ
also offers a recording and sound buyer’s guide.
Electronic Musician (home and project studio recording), www
.emusician.com; Electronic Musician offers a supplement called
“Desktop Music Production Guide.”
Tape Op (home and project studio creative recording), www
.tapeop.com.
Future Music (home, project and pro recording), www
.futuremusicmag.com.
Sound On Sound (Britain’s premier recording publication), www
.soundonsound.com.
In those magazines are ads related to recording products and services.
PRO AUDIO MAGAZINES
Journal of the Audio Engineering Society (JAES), pro audio engineering. Scholarly. With its journal, conventions, workshops, and
local chapters, the Audio Engineering Society is a tremendous
resource, www.aes.org.
Pro Audio Review (reviews of pro audio equipment), www.pro
audioreview.com.
Live Sound (concert sound reinforcement), www.livesoundint.com.
Church Production (audio for houses of worship), www.church
production.com.
Where to Learn More APPENDIX D
Technologies for Worship (audio for houses of worship), www
.tfwm.com.
Church and Worship Technology (audio for houses of worship),
http://churchandworship.com.
CONSUMER AUDIO MAGAZINES
Sound and Vision (consumer audio and video), www.soundand
visionmag.com.
Stereophile (high-end audio), www.stereophile.com.
The Absolute Sound (high-end audio), www.avguide.com/
the-absolute-sound.
GUIDES, BROCHURES, AND OTHER LITERATURE
Microphone application guides are available from:
Crown International, www.crownaudio.com.
Shure Inc., www.shure.com.
Countryman Associates Inc., www.countryman.com.
AKG Acoustics Inc., www.akg.com.
Sennheiser Electronic Corp., www.sennheiserusa.com.
Audio-Technica U.S. Inc., www.audio-technica.com.
Neumann/USA, www.neumannusa.com.
Schoeps Mikrofone, www.schoeps.de.
Danish Pro Audio, www.dpamicrophones.com (be sure to check
out their Microphone University)
You can find valuable information in user manuals and free sales
literature provided by manufacturers of recording equipment. Ask
your equipment dealer for manufacturers’ phone numbers and Web
site URLs.
GUIDES TO RECORDING SCHOOLS
The Audio Engineering Society offers a directory of educational programs at www.aes.org/education/directory.cfm.
Each July issue of Mix magazine contains a comprehensive directory
of recording schools, seminars, and programs. Universities and colleges in most major cities have recording-engineering courses.
543
544
APPENDIX D Where to Learn More
An index of recording schools is at www.modrec.com/schools/.
Two recent programs are www.recordingconnection.com, and the
Bachelor of Science in Music Technology at the University of Saint
Francis, School of Creative Arts in Fort Wayne, Indiana. For information call 260-434-3279.
The New Ears/Sound Schools book offers a list of recording schools.
The Web site is www.soundschools.com.
THE INTERNET
A great place to ask questions, besides magazines, is on the Internet.
In Google, select Groups, then type in rec.audio.pro. You can ask
questions and get answers from pro engineers. You may get conflicting answers, because often there are many ways to do the same
thing. Also, some who reply are more expert than others. But you’ll
often find stimulating debates.
Some other valuable Web sites (all starting with www.) are
homerecording.com, hometracked.com, prorec.com, kvraudio
.com,
gearslutz.com,
recordbetteraudio.com,
recording.org,
recordproduction.com,
vintagesynth.com,
moultonlabs.com,
microphones.com, digido.com, prosoundweb.com, digifreq.com,
cdrfaq.org, harmonycentral.com, musicplayer.com (forums), and
audio-recording-center.com.
Also see pcrecording.com, audiogrid.com, digitalprosound.com,
dvdforum.org, recordproduction.com, coutant.org, microphone-data.
com, themicrophonevault.com, miclisteningroom.org, audiodirectory.
nl/, binaural.com, sweetwater.com (check out publications and
forums), cakewalk.com (forums), and www.solorb.com/dat-heads/.
If you lack good multitrack recordings with which to practice mixing, go to www.raw-tracks.com. There you can download individual
tracks in WAV or MP3 format, or purchase a CD of raw tracks.
Some online MIDI resources are midi.com, midiworld.com, and
harmony-central.com.
The Web sites of audio equipment manufacturers have support sections, online discussion groups, and FAQs with lots of information.
Search for the name and model number of equipment you have, or
are interested in.
Where to Learn More APPENDIX D
Do a Google search with the keywords audio, recording, recording
vocals, audio recording links, MIDI, digital audio, and so on. Also
search for FAQs on various audio topics. You’ll discover hundreds of
audio-related Web sites and links. In Google, you can search Groups
as well as Web sites.
RECORDING EQUIPMENT CATALOGS
Here are but a few catalogs from which to order recording gear:
American Musical Supply, www.americanmusical.com
BSW, www.bswusa.com
Guitar Center, www.guitarcenter.com
Musician’s Friend, www.musiciansfriend.com
Sam Ash, www.samash.com
Prodigy Pro, www.prodigy-pro.com
B&H Pro Audio, www.bhproaudio.com
Sweetwater Sound, www.sweetwater.com
The Woodwind and the Brasswind, www.wwandbw.com
EXPERIENCE
It’s the best teacher. Record all you can with any equipment you
have. You can buy a 4-track cassette recorder on eBay for $45, or
download free multitrack recording software. Then buy a cheap mic,
and practice mic techniques and overdubbing.
545
This page intentionally left blank
Glossary
A weighting
See Weighted.
AAC Abbreviation for Advanced Audio Code. A relatively new compressed audio file format, AAC offers better sound quality than
MP3, but with less storage space and bandwidth. It’s also used
with Digital Rights Management technology to help control the
copying and distributing of music.
AAF Abbreviation for Advanced Authoring Format; www
.aafassociation.org/html/technoinfo/index.html. An industry
standard for high-end exchange of video or audio project data.
A-B A listening comparison between two audio programs, or
between two components playing the same program, performed
by switching immediately from one to the other. The levels of the
two signals are matched. See also Spaced-pair.
AC-3 Same as Dolby Digital, a perceptual encoding scheme that
data-reduces the six surround channels of a 5.1 system to two
channels.
Accent microphone See Spot microphone.
Access jacks See Insert jacks.
Active combining network A combining network with gain. See
Combining network.
AES
Audio Engineering Society.
AES/EBU Also called IEC 988 Type 1. An interface format for digital signals, using a balanced 110-ohm mic cable terminated with
XLR-type connectors. See also S/PDIF.
547
548
GLOSSARY
AIFF Abbreviation for Audio Interchange File Format. A standard
Mac format for uncompressed digital audio files.
Alignment The adjustment of analog tape-head azimuth and of
tape-recorder circuitry to achieve optimum performance from the
particular type of tape being used.
Alignment tape A prerecorded tape with calibrated tones for alignment of an analog tape recorder.
Ambience Room acoustics, early reflections, and reverberation.
Also, the audible sense of a room or environment surrounding a
recorded instrument.
Ambience microphone A microphone placed relatively far from its
sound source to pick up ambience.
Amplitude, Peak On a graph of a sound wave, the sound pressure
of the waveform peak. On a graph of an electrical signal, the voltage of the waveform peak. The amplitude of a sound wave or signal as measured on a meter is 0.707 times the peak amplitude.
Analog-to-digital (A/D) converter A circuit that converts an analog audio signal into a stream of digital data (bitstream).
Anti-alias filter In an A/D converter, a lowpass filter that removes
all frequencies above 20 kHz before sampling to prevent audio
artifacts called aliasing.
Anti-imaging filter In a D/A converter, a lowpass filter that
smooths the voltage steps in the analog signal that was generated by translating digital numbers into analog voltages. The antiimaging filter recovers the waveform of the original analog signal.
ASIO Abbreviation for Audio Stream Input/Output. Steinberg’s
computer audio driver spec for Mac and Windows. ASIO has low
latency because it interfaces directly between a sound card and
the audio application software.
GLOSSARY
Assign To route or send an audio signal to one or more selected
channels.
ATRAC Abbreviation for Adaptive Transform Acoustic Coding. A
data compression scheme (used in the MiniDisc) that reduces by
5:1 the storage needed for digital audio. ATRAC is a perceptual
coding method that omits data deemed inaudible due to masking.
Attack The beginning of a note. The first portion of a note’s envelope in which a note rises from silence to its maximum volume.
Attack time In a compressor, the time it takes for gain reduction to
occur in response to a musical attack.
Attenuate
To reduce the level of a signal.
Attenuator In a mixer (or mixing console) input module, an adjustable resistive network that reduces the microphone signal level to
prevent overloading of the input transformer and mic preamplifier.
Audio interface A device that connects to a computer and converts
an audio signal into computer data for storage in memory or
on hard disk. The interface also converts computer data into an
audio signal. See Breakout box, I/O box, and Sound card.
Autolocate A recorder function that makes the tape or disk go to a
program address (counter time) at the press of a button.
Automated mixing A system of mixing in which a computer
remembers and updates mixer control settings and moves. With
this system, a mix can be performed and refined in several stages
and played back at a later date exactly as set up previously.
Auxiliary bus (aux bus) A bus or channel that contains a mix of the
aux-send signals of the input modules in a mixer. An aux bus is used
to send signals to an effects unit or monitor system. See Effects bus.
Auxiliary send (aux-send) A control in a mixer’s input module
used to send that module’s signal to an aux bus. The aux-send
549
550
GLOSSARY
level adjusts the amount of effects heard on an instrument, or
adjusts the loudness of that instrument in the monitor system.
A/V drive A hard-disk drive meant for audio/video use. It postpones thermal recalibration until the disk is inactive, preventing
data errors.
Azimuth In a tape recorder, the angular relationship between the
head gap and the tape path.
Azimuth alignment In an analog tape recorder, the mechanical
adjustment of the record or playback head to bring it into proper
alignment (90 degrees) with the tape path.
Back-timing A technique of cueing up the musical background or
a sound effect to a narration track so that the music or effect ends
simultaneously with the narration.
Baffled-omni A stereo miking arrangement that uses two ear-spaced
omnidirectional microphones separated by a disk, sphere, or foam.
Balance The relative volume levels of various tracks or instruments.
Balanced AC power AC power from a center-tapped power transformer. Instead of one 120 V line and one 0 V line, it has two 60 V
lines. They are in phase with each other and sum to 120 V. But
they are connected to the center-tap ground out of phase (one is
60 V; the other is ⫺60 V). Any hum and noise on the grounding
system cancel out.
Balanced line A cable with two conductors surrounded by a shield
in which each conductor is at equal impedance to ground. With
respect to ground, the conductors are at equal potential but opposite polarity; the signal flows through both conductors.
Bandpass filter In a crossover, a filter that passes a band or range
of frequencies but sharply attenuates or rejects frequencies outside the band.
Basic tracks Recorded tracks of rhythm instruments (bass, guitar,
drums, and sometimes keyboard).
GLOSSARY
Bass management A subwoofer/satellite crossover, usually part of
a surround receiver. Bass management routes frequencies above
about 100 Hz to the five full-range speakers, and routes frequencies below about 100 Hz from all six channels to the subwoofer.
It affects only what you monitor, not what is recorded. Bass management can be done by a surround-receiver circuit, a stand-alone
box, a special circuit in a subwoofer, or a software plug-in.
Bass trap An assembly that absorbs low-frequency sound waves in
the studio.
Bi-amplification (bi-amping) Driving a woofer and tweeter with
separate power amplifiers. An active crossover is connected ahead
of these power amplifiers.
Bias In tape-recorder electronics, an ultrasonic signal that drives
the erase head. This signal is also mixed with the audio signal
applied to the record head to reduce distortion.
Bidirectional microphone A microphone that is most sensitive to
sounds arriving from two directions—in front of and behind the
microphone. It rejects sounds approaching either side of the microphone. Sometimes called a cosine or figure-eight microphone
because of the shape of its polar pattern.
Binaural recording A 2-channel recording made with an omnidirectional microphone mounted near each ear of a human or
a dummy head, for playback over headphones. The object is to
duplicate the acoustic signal appearing at each ear.
Bit depth The number of bits (ones and zeros) making up a word in
a digital signal (such as 16- or 24-bits). Each word is a binary number that is the value of each sample. A sample is a measurement of
an analog waveform that is done several thousand times a second
during analog-to-digital conversion. High bit depth ⫽ long word
length ⫽ high resolution of the analog signal amplitude ⫽ high
sound quality. See Word length.
551
552
GLOSSARY
Blumlein array A stereo microphone technique in which two coincident bidirectional microphones are angled 90 degrees apart (45
degrees to the left and right of center).
Board See Mixing console.
Bouncing tracks A process in which two or more tracks are mixed,
and the mixed tracks are recorded on an unused track or tracks.
Then the original tracks can be erased, which frees them up for
recording more instruments.
Boundary microphone A microphone designed to be used on a
boundary (a hard reflective surface). The microphone capsule is
mounted very close to the boundary so that direct and reflected
sounds arrive at the microphone diaphragm in phase (or nearly
so) for all frequencies in the audible band.
Breakout box (I/O box) A group of audio input and output connectors in a chassis, which is wired to a sound card (PCI audio
interface) in a computer. Used to interface analog audio signals
(and often MIDI and digital signals) with a computer.
Breathing The unwanted audible rise and fall of background noise
that may occur with a compressor. Also called pumping.
Bulk tape eraser A large electromagnet used to erase a whole reel
of recording tape at once.
Bus A common connection of many different signals. An output of
a mixer or submixer. A channel that feeds a tape track, signal processor, or power amplifier.
Bus in An input to a program bus, usually used for effects returns.
Bus master In the output section of a mixing console, a potentiometer (fader or volume control) that controls the output level of
a bus. See also Group fader.
Bus out The output connector of a bus.
GLOSSARY
Bus trim A control in the output section of a mixing console that
provides variable gain control of a bus, used in addition to the
bus master for fine adjustment.
Buzz An unwanted edgy tone that sometimes accompanies audio,
containing high harmonics of 60 Hz (50 Hz in Europe).
Calibration See Alignment.
Capacitor An electronic component that stores an electric charge.
It is formed of two conductive plates separated by an insulator
called a dielectric. A capacitor passes AC but blocks DC.
Capacitor microphone See Condenser microphone.
Capstan In a tape-recorder transport, a rotating post that contacts
the tape (along with the pinch roller) and pulls the tape past the
heads at a constant speed during recording and playback.
Cardioid microphone A unidirectional microphone with side
attenuation of 6 dB and maximum rejection of sound at the rear
of the microphone (180 degrees off-axis). A microphone with a
heart-shaped directional pattern.
CD
See Compact disc.
CD-R Abbreviation for CD-Recordable. A recordable compact disc
that can’t be rewritten. Once recorded, it can’t be erased and reused.
CD-ROM A computer disk drive that plays computer data from
CD-ROM disks. The disks are read optically by a laser, as in a
compact disc player.
CD-RW Abbreviation for CD-Rewritable. A recordable compact disc
that can be rewritten. Once recorded it can be erased and reused.
Channel A single path of an audio signal. Usually, each channel
contains a different signal.
Channel assign
See Assign.
553
554
GLOSSARY
Chorus (1) A special effect in which a signal is delayed by 15 to
35 msec, the delayed signal is combined with the original signal,
and the delay is varied randomly or periodically. This creates a
wavy, shimmering effect. (2) The main portion of a song that is
repeated several times throughout the song with the same lyrics.
Clean Free of noise, distortion, overhang, leakage. Not muddy.
Clear Easy to hear, easy to differentiate. Reproduced with sufficient
high frequencies.
Clip To cut off or flatten the tops of a signal waveform, causing distortion. Clipping is caused by setting signal level too high in an
audio device or audio software. See also Region.
Coincident-pair A stereo microphone, or two separate microphones,
placed so that the microphone diaphragms occupy approximately
the same point in space. They are angled apart and mounted one
directly above the other.
Comb-filter effect The frequency response caused by combining a
sound with its delayed replica. The frequency response has a series
of peaks and dips caused by phase interference. The peaks and
dips resemble the teeth of a comb.
Combining amplifier An amplifier in a mixing console at which
the outputs of two or more signal paths are mixed together to
feed a single track of a recorder.
Combining network A resistive network in a mixing console at
which the outputs of two or more signal paths are mixed together
to feed a single track of a recorder.
Compact disc (CD) A read-only optical disc medium for storing
digital audio programs up to 80 minutes long. The compact disc
stores data in the form of a spiral groove of microscopic pits and
is read optically by a laser. CD’s digital audio format is 44.1 kHz
sampling rate and 16-bit word length.
GLOSSARY
Comping Recording several musical performances (takes) of a single
instrument or vocal on different tracks, and selecting the best
parts of each take to be played in order on a composite track during mixdown.
Complex wave A wave with more than one frequency component.
Composite track A track containing the best parts of several takes
of a musical-instrument or vocal performance.
Compression (1) The portion of a sound wave in which air molecules are pushed together, forming a region with higher than normal atmospheric pressure. (2) In signal processing, the reduction
in dynamic range or gain caused by a compressor. (3) Data compression or data reduction is an encoding scheme to reduce the
size of a data file by throwing away audio data deemed inaudible
because of masking. ATRAC, MP3, MLP, AAC, RealAudio, OGG,
and Microsoft Media are examples of compressed data formats.
Compression ratio (slope) In a compressor, the ratio of the
change in input level (in dB) to the change in output level (in
dB). For example, a 2:1 ratio means that for every 2 dB change in
input level, the output level changes 1 dB.
Compressor A signal processor that reduces dynamic range or gain
by means of automatic volume control. An amplifier whose gain
decreases as the input signal level increases above a preset point.
Condenser microphone A microphone that works on the principle
of variable capacitance to generate an electrical signal. The microphone diaphragm and an adjacent metallic disk (called a backplate) are charged to form two plates of a capacitor. Incoming
sound waves vibrate the diaphragm, varying its spacing to the
backplate, which varies the capacitance, which in turn varies the
voltage between the diaphragm and backplate.
Connector A device that makes electrical contact between a signalcarrying cable and an electronic device, or between two cables.
555
556
GLOSSARY
A device used to connect or hold together a cable and an electronic component so that a signal can flow from one to the other.
Console See Mixing console.
Contact pickup A transducer that contacts a musical instrument
and converts its mechanical vibrations into a corresponding electrical signal.
Control room The room in which the engineer controls and monitors the recording. It houses most of the recording hardware.
Controller surface A chassis with faders (and sometimes buttons
and knobs) that resembles a mixer, used to adjust virtual controls
that appear on-screen in computer recording software. Connected
to the computer by USB or FireWire, the controller surface might
include analog and digital input/output connectors and MIDI
connectors.
Convolution reverb (sampling reverb) A reverberation device or
plug-in which creates the reverb from impulse-response samples
(WAVE files) of real acoustic spaces, rather than from algorithms.
The resulting sound quality is very natural.
Crossover An electronic network that divides an incoming signal
into two or more frequency bands.
Crossover, active (electronic crossover) A crossover with amplifying components, used ahead of the power amplifiers in a bi- or
tri-amped speaker system.
Crossover frequency The single frequency at which both filters of
a crossover network are down 3 dB.
Crossover, passive A crossover with passive (non-amplifying) components, used after the power amplifier.
Crosstalk The unwanted transfer of a signal from one channel to
another. Crosstalk often occurs between adjacent tracks within a
GLOSSARY
record or playback head in a tape recorder, or between input
modules in a console.
Cue, cue send In a mixing-console input module, a control that
adjusts the level of the signal feeding the cue mixer that feeds a
signal to headphones in the studio.
Cue list See Edit decision list (EDL).
Cue mixer A submixer in a mixing console that takes signals from
cue sends (aux sends) as inputs and mixes them into a composite
signal that drives headphones in the studio.
Cuesheet Used during mixdown, a chronological list of mixingconsole control adjustments required at various points in the
recorded song. These points may be indicated by counter or ABStime readings.
Cue system A monitor system that allows musicians to hear themselves and previously recorded tracks through headphones.
Damping factor The ability of a power amplifier to control or
damp loudspeaker vibrations. The lower the amplifier’s output
impedance, the higher the damping factor.
DAT (R-DAT) A digital audio tape recorder that uses a rotating
head to record digital audio on tape. It is obsolete, having been
replaced by Flash memory recorders.
Data compression A data encoding scheme for reducing the
amount of data storage on a medium. Same as Data Reduction.
See Compression, ATRAC, and MP3.
DAW
dB
Abbreviation for digital audio workstation.
Abbreviation for decibel.
Dead Having very little or no reverberation.
Decay The portion of the envelope of a note in which the envelope
goes from maximum to some midrange level. Also, the decline in
level of reverberation over time.
557
558
GLOSSARY
Decay time See Reverberation time (RT 60).
Decibel The unit of measurement of audio level. Ten times the
logarithm of the ratio of two power levels. Twenty times the logarithm of the ratio of two voltages. dBV is decibels relative to 1
volt, dBu is decibels relative to 0.775 volt, dBv is decibels relative to 0.7746 volt, dBm is decibels relative to 1 milliwatt, dBFS is
decibels, full scale (all bits on or off), dBSPL is decibels of sound
pressure level, and dBA is decibels, A weighted. See Weighted.
Decoded tape A program on analog recording tape that is
expanded after being compressed by a noise-reduction system.
Such a program has normal dynamic range.
De-esser A signal processor or plug-in that removes excessive sibilance (“s” and “sh” sounds) by compressing high frequencies
around 3 to 10 kHz.
Delay The time interval between a signal and its repetition. A digital delay or a delay line is a signal processor that delays a signal
for a short time.
Delay compensation Adjusting the timing of a track that is processed by a plug-in, to make it in sync with non-processed tracks.
Plug-ins add latency (delay) to a track.
Demagnetizer (degausser) An electromagnet with a probe tip that
is touched to elements of an analog recorder tape path (such as
tape heads and tape guides) to remove residual magnetism.
Depth The audible sense
instruments. Instruments
to-reverberant sound are
recorded with a low ratio
ceived as being distant.
of nearness and farness of various
recorded with a high ratio of directperceived as being close; instruments
of direct-to-reverberant sound are per-
Design center The portion of fader travel (usually shaded), about
10 to 15 dB from the top, in which console gain is distributed for
optimum headroom and signal-to-noise ratio. During normal operation, each fader in use should be placed at or near design center.
GLOSSARY
Designation strip A strip of paper taped near console faders to
designate the instrument that each fader controls. Also called a
scribble strip.
Desk
The British term for mixing console.
Destructive editing In a digital audio workstation, editing that
rewrites the data on disk. A destructive edit can’t be undone
unless a copy of the original data is saved before the edit is done.
DI
Abbreviation for direct injection, recording with a direct box.
Diffusion An even distribution of sound in a room.
Digital audio An encoding of an analog audio signal in the form
of binary digits (ones and zeroes).
Digital audio workstation (DAW) A computer, audio interface,
and recording software that allows you to record, edit, and mix
audio programs entirely in digital form. A stand-alone DAW is a
digital multitrack recorder-mixer. Stand-alone DAWs include real
mixer controls; computer DAWS have virtual controls on-screen.
Digital recording A recording system in which the audio signal is
stored as binary digits (ones and zeroes).
Digital-to-analog (D/A) converter A circuit that converts a digital
audio signal into an analog audio signal.
Dim To reduce the monitor volume temporarily by a preset
amount so that you can carry on a conversation.
Direct box A device used for connecting an amplified instrument
directly to a mixer mic input. The direct box converts a highimpedance unbalanced audio signal into a low-impedance balanced audio signal.
Direct injection (DI) Recording with a direct box.
Directional microphone A microphone that has different sensitivity in different directions. A unidirectional or bidirectional
microphone.
559
560
GLOSSARY
Direct output, direct out An output connector following a mic
preamplifier, fader, and equalizer, used to feed the signal of one
instrument to one track of a multitrack recorder.
Direct sound Sound traveling directly from the sound source to
the microphone (or to the listener) without reflections. Also,
DirectSound is an audio driver for the Windows operating system,
intended as an enhancement to MultiMedia Extensions (MME).
DirectX (DX) A package of Windows audio, video, and gamecontroller drivers, also called multimedia application programming interfaces (APIs).
Distortion An unwanted change in the audio waveform, causing a
raspy or gritty sound quality. The appearance of frequencies in a
device’s output signal that weren’t in the input signal. Distortion
is caused by recording at too high a level, using improper
mixer settings, components failing, or vacuum tubes distorting.
(Distortion can be desirable, e.g., for an electric guitar.) In digital
recording, distortion called quantization error can also occur at
very low signal levels, where there aren’t enough bits to record the
signal accurately. See also Clip.
Dither Low-level noise added to a digital signal to reduce quantization distortion caused by truncating (removing) bits in a digital word. It’s a good idea to add dither to a 24-bit program just
before converting it to 16 bits for CD release.
Dolby Digital A perceptual coding method using AC-3 data compression, offering 6 discrete channels of digital surround sound.
Dolby Digital uses a lossy encoding process to reduce the bit rate
needed to transmit the six channels via a 2-channel bitstream.
The standard audio format for DVD-Video.
Dolby Pro Logic A surround decoder that decodes the two channels of Dolby-Surround-encoded programs back into four channels (left, center, right, surround).
Dolby Surround A matrix encoding system that combines four channels (left, center, right, surround) into two channels. The Dolby
GLOSSARY
Pro Logic decoder unfolds the two channels back into four. The
surround channel, which is mono and limited bandwidth, is reproduced over left and right surround speakers.
Dolby Tone A reference tone recorded at the beginning of a Dolbyencoded analog tape for alignment purposes.
Doubling A special effect in which a signal is combined with its
15- to 35-msec-delayed replica. This process mimics the sound of
two identical voices or instruments playing in unison. In another
type of doubling, two identical performances are recorded and
played back to thicken the sound.
Drop-frame For color video production, a mode of SMPTE timecode
that causes the timecode to match the clock on the wall. Once
every minute, frame numbers 00 and 01 are dropped, except every
10th minute.
Drop-out During playback of a tape recording, a momentary loss
of signal caused by separation of the tape from the playback head
by dust, tape-oxide irregularity, etc. During playback of a hard-disk
recording, a momentary loss of signal caused by a buffer memory
being emptied. Increasing buffer size usually prevents drop-outs.
Drum machine A device—hardware or software—that plays samples
of real drums and includes a sequencer to record rhythm patterns.
Dry Having no echo or reverberation. Referring to a close-sounding
signal that hasn’t yet been processed by a reverberation or delay
device or plug-in.
DSD Abbreviation for Direct Stream Digital. A process that encodes
a digital signal in a 1-bit (bitstream) format at a 2.8224 MHz sampling rate. Offers state-of-the-art sound quality; used in the Super
Audio CD.
DSP Abbreviation for Digital Signal Processing. Modifying a signal
in digital form by doing calculations on the numbers. DSP is used
for level changes, EQ, effects, and so on.
561
562
GLOSSARY
DTS Abbreviation for Digital Theater System. A perceptual coding method using data compression, offering 6 discrete channels
of digital surround sound. DTS uses a lossy encoding process
to reduce the bitrate needed to transmit the 6 channels via a
2-channel bitstream.
DVD Abbreviation for Digital Versatile Disc. A storage medium
the size of a compact disc that holds much more data. The DVD
stores video, audio, or computer data. DVD-RAM and DVD-R are
recordable; DVD-RW is rewritable, and DVD-A is audio only.
DVD-Audio A DVD intended mainly for audio programs. It can
use Dolby Digital or DTS encoded programs.
DX
See DirectX.
DXi Abbreviation for DirectX Instruments. Cakewalk’s standard
format for plug-in software synthesizers based on Microsoft
DirectX technology.
Dynamic microphone A microphone that generates electricity
when sound waves cause a conductor to vibrate in a stationary
magnetic field. The two types of dynamic microphones are moving coil and ribbon. A moving-coil microphone is usually called a
dynamic microphone.
Dynamic range The range of volume levels in a program from softest to loudest.
Earth ground A connection to moist dirt (the ground we walk on).
This connection is usually done via a long copper rod.
EASI Abbreviation for Enhanced Audio Streaming Interface. Emagic’s
sound card driver spec. EASI achieves low latency by interfacing the
sound card directly with the audio application software.
Echo A delayed repetition of a signal or sound. A sound delayed
50 msec or more that is combined with the original sound.
GLOSSARY
Echo chamber A hard-surfaced room, containing a widely separated
loudspeaker and microphone, used for creating reverberation.
Edit decision list (EDL) A list of program events in order, plus
their starting and ending times.
Editing The cutting and rejoining of magnetic tape to delete
unwanted material, to insert silent spaces, or to rearrange recorded
material into the desired sequence. Also, the same actions performed on a digital recording with a DAW, hard-disk recorder, or
MiniDisc recorder–mixer.
Editing block A metal block that holds magnetic tape during the
editing/splicing procedure.
Effects Interesting sound phenomena created by signal processors,
such as reverberation, echo, flanging, doubling, compression, or
chorus. See Sound effects.
Effects bus The bus that feeds effects devices (signal processors).
Effects loop A set of connectors in a mixer for connecting an external effects unit, such as a reverb or delay device. The effects loop
includes a send section and a receive section. See Effects send,
Effects return.
Effects return (aux return) In the output section of a mixing console,
a control that adjusts the amount of signal received from an effects
unit. Also, the connectors in a mixer to which you connect the
effects-unit output signal. They might be labeled “bus in” instead.
The effects-return signal is mixed with the program bus signal.
Effects send (aux send) In an input module of a mixing console, a
control that adjusts the amount of signal sent to an effects device,
such as a reverberation or delay unit. Also, the connector in a
mixer that you connect to the input of an effects unit. The effectssend or aux-send control adjusts the amount of effects heard on
each instrument.
563
564
GLOSSARY
Efficiency In a loudspeaker, the ratio of acoustic power output to
electrical power input.
EIA
Electrical Industries Association.
Electret-condenser microphone A condenser microphone in which
the electrostatic field of the capacitor is generated by an electret—a
material that permanently stores an electrostatic charge.
Electrostatic field The force field between two conductors charged
with static electricity.
Electrostatic interference The unwanted presence of an electrostatic hum field in signal conductors.
Encoded tape An analog tape containing a signal compressed by a
noise-reduction unit.
End-addressed Referring to a microphone whose main axis of
pickup is perpendicular to the front of the microphone. You aim
the front of the mic at the sound source. See Side-addressed.
Envelope The rise and fall in volume of one note. The envelope
connects successive peaks of the waves that make up the note.
Each harmonic in the note might have a different envelope. A volume envelope in a DAW track is a graph of track volume (fader
setting) versus time.
Equalization (EQ) The adjustment of frequency response to alter
the tonal balance or to attenuate unwanted frequencies.
Equalizer A circuit (usually in each input module of a mixing console, or in a separate unit) that alters the frequency spectrum of a
signal passed through it.
Erase To remove an audio signal from magnetic tape or disk by
applying an ultrasonic varying magnetic field to randomize the
magnetization of the magnetic particles on the tape or disk.
Erase head A head in a tape recorder or hard drive that erases the
signal on tape or disk.
GLOSSARY
Expander (1) A signal processor that increases the dynamic range of
a signal passed through it. (2) An amplifier whose gain decreases
as its input level decreases. When used as a noise gate, an expander
reduces the gain of low-level signals to reduce noise between notes.
Fade-out To gradually reduce the volume of the last several seconds
of a recorded song, from full level down to silence, by slowly pulling down the master fader, or by selecting a fade-out process in a
digital editing program.
Fader A linear or sliding potentiometer (volume control), used to
adjust signal level.
Feed (1) To send an audio signal to some device or system. (2) An
output signal sent to some device or system.
Feedback (1) The return of some portion of an output signal to
the system’s input. (2) The squealing sound you hear when a PA
system microphone picks up its own amplified signal through a
loudspeaker.
Feed reel The left-side reel on an analog tape recorder that unwinds
during recording or playback.
Filter (1) A circuit that sharply attenuates frequencies above or
below a certain frequency. Used to reduce noise and leakage
above or below the frequency range of an instrument or voice. (2)
A MIDI filter removes selected note parameters.
FireWire A standard protocol for high-speed transfer of data
between digital devices. Also called IEEE 1394.
Fixed point In digital signal processing, fixed-point calculations
use numbers with a fixed number of digits after the decimal
point. See Floating Point.
Flanging A special effect in which a signal is combined with its
delayed replica, and the delay is varied between 0 and 20 msec.
565
566
GLOSSARY
A hollow, swishing, ethereal effect like a variable-length pipe, or
like a jet plane passing overhead. A variable comb filter produces
the flanging effect.
Fletcher-Munson effect Named after the two people who discovered it, the psychoacoustical phenomenon in which the subjective
frequency response of the ear changes with program level. Because
of this effect, a program played at a lower volume than the original
level subjectively loses low- and high-frequency response.
Float To disconnect from ground.
Floating point In digital signal processing, calculations using numbers with a fixed number of bits for the mantissa and the remaining bits for the exponent. 32-bit “float” uses a 24-bit mantissa
and an 8-bit exponent. It’s almost impossible to clip an audio
signal in 32-bit floating-point mixing software. When the clip
light flashes, the signal isn’t clipping in the DAW, but it would be
clipped if it were exported to a CD, to an audio interface, or to a
24-bit fixed-point file. See Fixed point.
Flutter
A rapid periodic variation in tape speed.
Flutter echoes A rapid series of echoes that occurs between two
parallel walls.
Flux Magnetic lines of force.
Fluxivity The measure of the flux density of a magnetic recording
tape, per unit of track width.
Fly-in (lay-in) To copy part of a recorded track onto another
recorder, then re-record that copy back onto the original multitrack
tape in a different part of the song, in sync with other recorded
tracks. For example, copy the vocal track from the first chorus of
the song onto an external DAW or sampler. Re-record (fly-in) that
copy onto the multitrack tape at the second chorus. Then the first
and second choruses have identical vocal performances.
GLOSSARY
Foldback (FB) See Cue system.
Frequency The number of cycles per second of a sound wave or
an audio signal, measured in hertz (Hz). A low frequency (for
example, 100 Hz) has a low pitch; a high frequency (for example,
10,000 Hz) has a high pitch.
Frequency response (1) The range of frequencies that an audio
device will reproduce at an equal level (within a tolerance, such
as ⫾3 dB). (2) The range of frequencies that a device (mic, human
ear, etc.) can detect.
Full-duplex Describing a sound card that can record and play back
simultaneously. The card works on two DMA channels.
Full track A single tape track recorded across the full width of an
analog tape.
Fundamental The lowest frequency in a complex wave.
FX
Abbreviation for effects.
Gain Amplification. The ratio, expressed in decibels, between
the output voltage and the input voltage, or between the output
power and the input power.
Gain staging Setting all the gain controls in a signal path to get
the cleanest signal. At any stage in the signal path, the signal level
should be set well above the noise floor but below clipping.
Gap In a tape-recorder head, the thin break in the electromagnet
that contacts the tape. In an audio program, the space or silence
between songs.
Gate (1) To turn off a signal when its amplitude falls below a preset value. (2) The signal-processing device used for this purpose.
See also Noise gate.
Gated reverb Reverberation with the reverberant “tail” cut off
before it fades out.
567
568
GLOSSARY
General MIDI file (GM file) A MIDI file containing a standard set
of musical instrument sounds. A General MIDI file produces the
same sounds on any MIDI-capable instrument that supports the
GM spec.
Generation A copy of a tape or a bounce of a track. A copy of the
original master recording is a first generation tape. A copy made
from the first generation tape is a second generation, and so on.
Generation loss The degradation of signal quality (the increase in
noise and distortion) that occurs with each successive generation
of an analog tape recording.
Gobo A moveable partition used to prevent the sound of an instrument from reaching another instrument’s microphone. Short for
go-between.
Graphic equalizer An equalizer with a horizontal row of faders; the fader-knob positions indicate graphically the frequency
response of the equalizer. Usually used to equalize monitor
speakers for the room they are in. Sometimes used for complex
EQ of a track.
Ground The zero-signal reference point for a system of audio
components.
Ground bus A common connection to which equipment is
grounded, usually a heavy copper plate.
Grounding Connecting pieces of electronic equipment to ground.
Proper grounding ensures that there is no voltage difference
between equipment chassis. An electrostatic shield needs to be
grounded to be effective.
Ground loop (1) A loop or circuit formed of ground leads. (2) The
loop formed when components are connected together via two
ground paths—the connecting-cable shield and the power ground.
Ground loops cause hum and should be avoided.
GLOSSARY
Group (1) To select several faders to make them act in unison. For
example, select all the faders for the drum tracks to so that you
can adjust the overall level of the drums by pushing one fader.
(2) To assign the output of several input modules to a single
group or bus, whose level is controlled by a single group fader. For
example, assign all the input modules of the drum mics to a single
“drums” group. (3) A bus or channel in a mixer that contains the
signals from several input modules. For example, a drums group is
a group of all the signals of the drum-set mics. See Submix.
Group fader (submaster fader) In the output section of a mixing
console, a potentiometer (fader or volume control) that controls
the output level of a bus or group.
GSIF Abbreviation for GigaSampler Interface. Nemesys’ Windows
sound card driver that achieves low latency by interfacing directly
between the sound card and the audio application software.
Guard band The spacing between tracks on a multitrack analog
tape or tape head, used to prevent crosstalk.
Half-track A tape track recorded across approximately half the width
of a tape. A half-track recorder usually records two such tracks
simultaneously in the same direction to make a stereo recording.
Hard disk A random-access storage medium for computer data. A
hard-disk drive contains a stack of magnetically coated hard disks
that are read by, and written to by, an electromagnetic head.
Hard-disk recorder (hard-drive recorder) A device dedicated
to recording digital audio on a hard-disk drive. A hard-disk
recorder–mixer includes a built-in mixer.
Harmonic An overtone whose frequency is a whole-number multiple of the fundamental frequency.
Harmonizer A signal processor that provides a wide variety of
pitch-shifting and delay effects.
569
570
GLOSSARY
HD
Abbreviation for hard-disk drive.
Head An electromagnet in a tape recorder that either erases the
audio signal on tape, records a signal on tape, or plays back a signal that is already on tape. A hard disk drive also has heads with
similar functions.
Head gap See Gap.
Headphones A head-worn transducer that covers the ears and converts electrical audio signals into sound waves.
Headroom The safety margin, measured in decibels, between the
signal level and the maximum undistorted signal level. In a tape
recorder, the dB difference between standard operating level (corresponding to a 0 VU reading) and the level causing 3% total
harmonic distortion. High-frequency headroom increases with
analog tape speed.
Hertz (Hz) Cycles per second, the unit of measurement of
frequency.
Highpass filter A filter that passes frequencies above a certain frequency and attenuates frequencies below that same frequency. A
low-cut filter.
Hiss A noise signal containing all frequencies, but with greater
energy at higher octaves. Hiss sounds like wind blowing through
trees. It is usually caused by random signals generated by microphones, electronics, and magnetic tape.
Host A DAW recording program that supports plug-ins. See Plug-in.
Hot (1) A high recording level causing slight distortion, may be
used for special effect. (2) High average level on a CD making
it relatively loud, produced by peak limiting and normalization,
or by compression and normalization. (3) A condition in which
a chassis or circuit has a potentially dangerous voltage on it.
(4) Referring to the conductor in a microphone cable that has a
GLOSSARY
positive voltage on it at the instant that sound pressure moves the
diaphragm inward.
Hum An unwanted low-pitched tone (60 Hz or 50 Hz and its harmonics) heard in the monitors. The sound of interference generated
in audio circuits and cables by AC power wiring. Hum pickup is
caused by such things as placing audio cables near power cables or
transformers, faulty grounding, poor shielding, and ground loops.
Hypercardioid microphone A directional microphone with a polar
pattern that has 12 dB attenuation at the sides, 6 dB attenuation
at the rear, and two nulls of maximum rejection at 110 degrees
off-axis.
Image An illusory sound source located somewhere around the listener. An image is generated by two or more loudspeakers. In a
typical stereo system, images are located between the two stereo
speakers.
Impedance The opposition of a circuit to the flow of alternating
current. Impedance is the complex sum of resistance and reactance. Abbreviated as Z.
Input The connection going into an audio device. In a mixer or
mixing console, a connector for a microphone, line-level device,
or other signal source.
Input attenuator See Attenuator.
Input module In a mixing console, the set of controls affecting a
single input signal. An input module usually includes an attenuator (trim), fader, equalizer, aux sends, and channel-assign buttons.
Input/output (I/O) console (in-line console) A mixing console
arranged so that input and output sections are aligned vertically.
Each module (other than the monitor section) contains one input
channel and one output channel.
Input section The row of input modules in a mixing console.
571
572
GLOSSARY
Insert jacks Two jacks (send and return) in a console input module
or output module that allow access to points in the signal path,
usually for connecting a compressor. Plugging into the insert jacks
breaks the signal flow and allows you to insert a signal processor
or recorder in series with the signal. In many mixers, a single insert
jack has both send and return terminals. Also called Access jacks.
Intersample peak: An analog signal produced by a digital-toanalog converter that momentarily exceeds the level of the
digital samples from which it was converted.
I/O
Referring to Input and Output connectors.
I/O box A Breakout Box type of audio interface.
Jack A female or receptacle-type connector for audio signals into
which a plug is inserted.
Keyboard workstation Several MIDI components in one chassis—
a keyboard, a sample player, a sequencer, and perhaps a synthesizer and disk drive.
Kilo A prefix meaning one thousand. Abbreviated k.
Latency The signal delay through an A/D, D/A converter, through
a software program, or through a computer operating system.
Monitoring latency is the delay between the time when a musician plays a note and when she hears the monitored signal of that
note. Latency can make you play out of sync during overdubs.
Lay-in
See Fly-in.
Leadering The process of splicing leader tape between program
selections on analog tape.
Leader tape Plastic or paper tape without an oxide coating, used for
a spacer between takes (for silence between songs) on analog tape.
Leakage The overlap of an instrument’s sound into another instrument’s microphone. Also called bleed or spill.
GLOSSARY
LEDE Abbreviation for Live-End/Dead-End, a type of control room
acoustic treatment in which the front half of the control room prevents early reflections to the mixing position, while the back half
of the control room reflects diffused sound to the mixing position.
LED indicator A recording-level indicator using one or more lightemitting diodes (LEDs).
Level The degree of intensity of an audio signal—the voltage,
power, or sound pressure level. The original definition of level is
the power in watts.
Level setting In a recording system, the process of adjusting the
input-signal level to obtain maximum level on the recording
medium without distortion. A VU meter, LED meter, or other
indicator shows recording level.
Lightpipe An Alesis connection protocol that transfers 8 digital
audio channels at once over a TOSLINK fiber-optic cable.
Limiter A signal processor whose output is constant above a preset input level. A compressor with a compression ratio of 10:1
or greater, with the threshold set just below the point of distortion of the following device. Used to prevent distortion of attack
transients or peaks. Also used to decrease the level of signal
peaks so that the average level of a program can be raised. See
Look-ahead.
Line level In balanced professional recording equipment, a signal
whose level is approximately 1.23 volts ⫹4 dBm). In unbalanced
equipment (most home hi-fi or semi-pro recording equipment), a
signal whose level is approximately 0.316 volt (⫺10 dBV).
Live (1) Having audible reverberation. (2) Occurring in real time,
in person.
Live recording A recording made at a concert. Also, a recording made of a musical ensemble playing all at once, rather than
overdubbing.
573
574
GLOSSARY
Localization The ability of the human hearing system to tell the
direction of a real or illusory sound source.
Locate (Autolocate) A recorder function that makes the tape or
disk head go to a specified program address (counter time) at the
press of a button.
Look-ahead A limiter plug-in with the “look-ahead” feature checks
the upcoming audio for peaks. Audio enters a look-ahead buffer,
and the limiter measures that audio signal and reacts quickly to
reduce the incoming peaks.
Loop In a sampling program, to play the sustain portion of a
sound’s envelope repeatedly. Also, a repeated rhythmic or musical
pattern.
Loudspeaker A transducer that converts electrical energy (the signal) into acoustical energy (sound waves).
Lowpass filter A filter that passes frequencies below a certain frequency and attenuates frequencies above that same frequency. A
high-cut filter.
M
Abbreviation for mega, or one million (as in megabytes). Small
m (m) is an abbreviation for milli, or one thousandth.
Magnetic recording tape A recording medium made of magnetic
particles (usually ferric oxide) suspended in a binder and coated
on a long strip of thin plastic (usually Mylar).
Magneto-optical drive (MO drive) A drive that stores data (such as
audio) on a 3.5-inch rewritable magneto-optical disk. The drive uses
a laser and magnetic head to write data, and a laser to read data.
Mask To hide or cover up one sound with another sound. To
make a sound inaudible by playing another sound along with it.
Masking is used in many data reduction schemes.
Master A completed tape or CD used to generate tape copies or
compact discs.
GLOSSARY
Master fader A volume control that affects the level of all program
busses simultaneously. It’s the last stage of gain adjustment before
the 2-track recorder.
MD
MDM
Abbreviation for MiniDisc.
Abbreviation for Modular Digital Multitrack.
Memory A group of integrated circuit chips used to store digital
data temporarily or permanently (such as an audio signal in digital format).
Memory recorder A device that records audio on a memory
chip, such as Compact Flash. Usually the recorded audio can be
uncompressed WAV files or compressed MP3 files.
Memory rewind A tape-recorder function that rewinds the tape to
a preset tape-counter position.
Meter A device that indicates voltage, resistance, current, or signal
level.
MFX The architecture or protocol for MIDI effects. See MIDI effects.
MIC
An abbreviation for microphone.
Mic level The level or voltage of a signal produced by a microphone, typically 2 millivolts.
Mic preamp See Preamplifier.
Microphone A transducer or device that converts an acoustical signal (sound) into a corresponding electrical signal.
Microphone techniques The selection and placement of microphones to pick up sound sources.
MIDI Abbreviation for Musical Instrument Digital Interface, a specification for a connection between synthesizers, drum machines,
575
576
GLOSSARY
and computers that allows them to communicate with and/or
control each other.
MIDI channel A route for transmitting and receiving MIDI signals.
Each channel controls a separate MIDI musical instrument or synth
patch. Up to 16 channels can be sent on a single MIDI cable.
MIDI clock A timing reference that sets a common tempo for all
the equipment. The MIDI clock is a series of bytes in the MIDI
data stream that conveys timing information. The clock is like a
conductor’s baton movements, keeping all the performers in sync
at the same tempo. The clock bytes are added to the MIDI performance information in the MIDI signal. The clock signal is 24,
48, or 96 pulses per quarter note (ppq). That is, for every quarter
note of the performance, 24 or more clock pulses (bytes) are sent
in the MIDI data stream.
MIDI controller A musical performance device (keyboard, drum
pads, breath controller, etc.) that outputs a MIDI signal designating note numbers, note on, note off, and so on.
MIDI/digital audio software Software that combines a MIDI
sequencer with a multitrack digital audio recorder/editor.
MIDI effects (MFX) Non-audio processes applied to MIDI signals, such as an arpeggiator, echo/delay, chord analyzer, quantize,
transpose MIDI, event filter, or velocity change. They can be used
as real-time, nondestructive plug-ins in MIDI tracks.
MIDI in
A connector in a MIDI device that receives MIDI messages.
MIDI interface A circuit that plugs into a computer and converts
MIDI data into computer data for storage in memory or on hard
disk. The interface also converts computer data into MIDI data.
MIDI out A connector in a MIDI device that transmits MIDI
messages.
MIDI thru A connector in a MIDI device that duplicates the MIDI
information at the MIDI-IN connector. Used to connect another
MIDI device in the series.
GLOSSARY
MIDI Timecode (MTC) See MTC.
Mid-side A coincident-pair stereo microphone technique using a
forward-facing unidirectional, omnidirectional, or bidirectional
mic and a side-facing bidirectional mic. The microphone signals
are summed and differenced to produce right- and left-channel
signals.
Mike
To pick up with a microphone.
milli A prefix meaning one thousandth, abbreviated m.
MiniDisc (MD) A rewritable, magneto-optical storage medium that is
read by a laser. It resembles a compact disc in a 2.5-inch square housing. MD recorders use a data compression scheme called ATRAC.
Mix (1) To combine two or more different signals into a common
signal. (2) A control on an effects processor that varies the ratio
between the dry (unprocessed) signal and the processed signal.
Mixdown The process of playing recorded tracks through a mixer
and mixing them to two stereo channels for recording on a twotrack recorder. Also applies to a surround-sound mixdown to 6 or
8 channels.
Mixer A device or software that mixes or combines audio signals
and controls the relative levels of the signals.
Mixing console A large mixer with additional functions such as
equalization or tone control, pan pots, monitoring controls, solo
functions, channel assigns, and control of signals sent to external
signal processors.
MLP Abbreviation for Meridian Lossless Packing. Used in DVDAudio discs, a data-reduction method that compresses six fullrange channels of 24-bit, 96 kHz audio without data loss.
MMC Abbreviation for MIDI Machine Control. A set of MIDI commands by which one device can control another. Some commands include Start, Stop, and Locate. MMC doesn’t include sync
information, but MTC does.
577
578
GLOSSARY
Modeler A device or software that simulates the sound of a microphone, guitar amp, or room.
MO drive See Magneto-optical drive.
Modular digital multitrack (MDM) A multitrack tape recorder that
records 8 tracks digitally on a videocassette. Several 8-track modules
can be linked together to add more tracks in sync. Two examples
of MDMs are the Alesis ADAT and TASCAM DA-88. This device is
obsolete and has been replaced by multitrack hard-disk recorders.
Monaural Referring to listening with one ear. Often incorrectly
used to mean monophonic.
Monitor A loudspeaker in a control room, or headphones, used
for judging sound quality. Also, a video display screen used with
a computer.
Monitoring Listening to an audio signal with a monitor.
Mono, monophonic (1) Referring to a single channel of audio.
A monophonic program can be played over one or more loudspeakers, or one or more headphones. (2). Describing a synthesizer that plays only one note at a time (not chords).
Mono-compatible A characteristic of a stereo program, in which
the program channels can be combined to a mono program without altering the frequency response or balance. A mono-compatible stereo program has the same frequency response in stereo or
mono because there is no delay or phase shift between channels
to cause phase interference.
Moving-coil microphone A dynamic microphone in which the
conductor is a coil of wire moving in a fixed magnetic field. The
coil is attached to a diaphragm that vibrates when struck with
sound waves. Usually called a dynamic microphone.
MP3 (MPEG Level-1 Layer-3) A data compression format for audio.
In an MP3 file (.mp3), the data has been compressed or reduced to
GLOSSARY
one-fourth of its original size or less. Compressed files take up less
memory, so they download faster. You download MP3 files to your
hard drive, then listen to them. MP3 audio quality at a 128 kbps rate
is nearly the same as that of CDs (depending on source material).
MP3PRO A data-compression format for audio. MP3Pro is an
improvement over MP3. Songs encoded at 64 kbps with MP3Pro
are said to sound as good as songs encoded at 128 kbps with MP3.
MP3Pro offers faster downloads and nearly double the amount of
music you can put on a Flash memory player. MP3 and MP3Pro
files are compatible with each other’s players, but an MP3Pro player
is needed to hear MP3Pro’s improvement in sound quality.
M-S RECORDING See Mid-side.
MTC Abbreviation for MIDI Timecode. A form of timecode transmitted over a MIDI cable, used for synchronizing MIDI devices, in
the form hours:minutes:seconds:frames. MIDI messages that provide SMPTE timecode information, used to sync a sequencer or
DAW with other devices that can synchronize to MTC. MTC also
syncs those devices to a tape recording that is striped with SMPTE
timecode. Unlike SMPTE timecode, MTC isn’t sample-accurate.
MTC isn’t affected by tempo, but MIDI clock is.
Muddy Unclear sounding; having excessive leakage, reverberation,
or overhang.
Multi-effects processor See Multiprocessor.
Multiple-D Microphone A directional microphone that has multiple sound-path lengths between its front and rear sound entries.
This type of microphone has minimal proximity effect.
Multiprocessor A signal processor that can perform several different signal-processing functions.
Multitimbral In a synthesizer, the ability to produce two or more
different patches or timbres at the same time.
Multitrack Referring to a recorder with more than two tracks.
579
580
GLOSSARY
Mute To turn off an input signal on a mixer by disconnecting the
input-module output from channel assign. During mixdown, the
mute function is used to reduce noise and leakage during silent portions of tracks, or to turn off unused performances. During recording, mute is used to turn off tracks that you don’t want to hear.
Near coincident A stereo microphone technique in which two
directional microphones are angled apart symmetrically on either
side of center and spaced a few inches apart horizontally.
Nearfield monitoring A monitor-speaker arrangement in which the
speakers are placed very near the listener (usually just behind the
mixing console) to reduce the audibility of control-room acoustics.
Noise Unwanted sound, such as hiss from electronics or tape. An
audio signal with an irregular, non-periodic waveform.
Noise gate A gate used to reduce or eliminate noise or leakage
between notes.
Noise-reduction system A signal processor used to reduce analog
tape hiss (and sometimes print-through) caused by the recording
process. The Dolby noise-reduction system compresses the high
frequencies during recording and expands them in a complementary way during playback.
Noise shaping Filtering the noise added in dithering to make the
noise less audible. Usually the filter reduces the level in the upper
midrange and increases the level at high frequencies.
Nondestructive editing In a digital audio workstation, editing
done by changing pointers (location markers) to information on
the hard disk. A nondestructive edit can be undone.
Nonlinear (1) Referring to a storage medium in which any data
point can be accessed or read almost instantly in a random fashion, rather than sequentially. Examples are a hard disk, compact
disc, and MiniDisc. See Random access. (2) Referring to an audio
device that is distorting the signal.
Normalize To raise the level of a digital audio signal so that the
highest peak in the program is at the highest level allowed by the
GLOSSARY
recording. For example, in a normalized 16-bit recording, the
highest peak in the program has all 16 bits on: the highest possible level in a 16-bit recording.
Octave The interval between any two frequencies where the upper
frequency is twice the lower frequency.
Off-axis Not exactly in front of a microphone or loudspeaker.
Off-axis coloration In a microphone, the deviation from the onaxis frequency response that sometimes occurs at angles off the
axis of the microphone. The coloration of sound (alteration of
tone quality) for sounds arriving off-axis to the microphone.
OGG The file extension for Ogg Vorbis, a data-reduction encoding
scheme.
OMF A file format for the transfer of digital data between DAW
software. You might record a project with Cakewalk Sonar, then
export it as an OMF file to be imported into a Pro Tools session.
Omnidirectional microphone A microphone that is equally sensitive to sounds arriving from all directions.
On-location recording A recording made outside the studio, in a
room or hall where the music usually is performed or practiced.
Open tracks On a multitrack recorder, tracks that haven’t been
used yet, or have already been bounced and are available for use.
ORTF Named after the French broadcasting network Office de
Radio-diffusion Television Française. A near-coincident stereo mic
technique that uses two cardioid mics angled 110 degrees apart
and spaced 17 cm horizontally.
Outboard equipment Signal processors that are external to the
mixing console.
Output A connector in an audio device from which the signal
comes and feeds successive devices.
581
582
GLOSSARY
Outtake A take, or section of a take, that is to be removed or not used.
Overdub To record a new musical part on an unused track in synchronization with previously recorded tracks.
Overhang The continuation of a signal at the output of a device
after the input signal has ceased. Sometimes called ringing.
Overload The distortion that occurs when an applied signal
exceeds a system’s maximum input level.
Oversampling In A/D conversion, oversampling is sampling an
audio signal at a higher rate than needed to reproduce the highest frequency in the signal. For example, sampling an audio signal at 8 times 44.1 kHz is called “8 ⫻ oversampling.” This process
is followed by a digital lowpass filter and a gentle-slope analog
anti-alias filter. The result is less phase shift and smoother sound
compared to a steep, “brick-wall” analog filter used alone. See
Anti-alias filter. Oversampling also can be applied in D/A conversion. Oversampling D/A converters raise the effective sampling
rate by interpolating new samples between the existing samples.
Overtone In a complex wave, a frequency component that is higher
than the fundamental frequency.
Pad
See Attenuator.
Pan pot Shortened term for panoramic potentiometer. In each
input module in a mixing console, a control that divides a signal
between two channels in an adjustable ratio. By doing so, a pan
pot controls the location of a sonic image between a stereo pair
of loudspeakers.
Parametric equalizer An equalizer with continuously variable parameters, such as frequency, bandwidth, and amount of boost or cut.
Patch (1) To connect one piece of audio equipment to another
with a cable. (2) A setting of synthesizer parameters to achieve a
sound with a certain timbre.
GLOSSARY
Patch bay (Patch panel) An array of connectors, usually in a rack,
to which equipment inputs and outputs are wired. A patch bay
makes it easy to interconnect various pieces of equipment in a
central, accessible location.
Patch cord A short length of cable with a phone plug on each end,
used for signal routing in a patch bay.
PCM Abbreviation for Pulse Code Modulation, a method of analog-to-digital conversion in which the instantaneous amplitude
of an analog waveform is measured or sampled several thousand
times a second, and each measurement is assigned a binary value
of a certain number of bits (ones and zeroes).
PDM Abbreviation for Pulse Density Modulation, a method of
analog-to-digital conversion in which the instantaneous amplitude of an analog waveform is coded as variations in the average
number of fixed-width pulses per unit of time.
Peak On a graph of a sound wave or signal, the highest point in the
waveform. The point of greatest voltage or sound pressure in a cycle.
Peak amplitude See Amplitude, Peak.
Peaking equalizer An equalizer that provides maximum cut or
boost at one frequency, so that the resulting frequency response
of a boost resembles a mountain peak.
Peak program meter (PPM) A meter that responds fast enough to
closely follow the peak levels in a program.
Period The time between the peak of one wave and the peak of the
next. The time between corresponding points on successive waves.
Period is the inverse of frequency.
Personal studio A minimal group of recording equipment set up
for one’s personal use, usually using a 4-track cassette or memory
recorder–mixer. Also, a simple 4-track recorder–mixer for one’s
personal use.
583
584
GLOSSARY
Perspective In the reproduction of a recording, the audible sense
of distance to the musical ensemble, the point of view. A close
perspective has a high ratio of direct sound to reverberant sound;
a distant perspective has a low ratio of direct sound to reverberant
sound.
PFL
Abbreviation for pre-fader listen. See also Solo.
Phantom power A DC voltage (usually 12 to 48 volts) applied to
microphone signal conductors to power condenser microphones.
Phase The degree of progression in the cycle of a wave, where one
complete cycle is 360 degrees.
Phase cancellation, Phase interference The cancellation of certain
frequency components of a signal that occurs when the signal
is combined with its delayed replica. At certain frequencies, the
direct and delayed signals are of equal level and opposite polarity
(180 degrees out of phase), and when combined, they cancel out.
The result is a comb-filter frequency response having a periodic
series of peaks and dips. Phase interference can occur between
the signals of two microphones picking up the same source at different distances, or can occur at a microphone picking up both a
direct sound and its reflection from a nearby surface.
Phase shift The difference in degrees of phase angle between corresponding points on two waves. If one wave is delayed with respect
to another, there is a phase shift between them of 2piFT, where
pi ⫽ 3.14, F ⫽ frequency in Hz, and T ⫽ delay in seconds.
Phasing A special effect in which a signal is combined with its
phase-shifted replica to produce a variable comb-filter effect. See
also Flanging.
Phone plug A cylindrical, coaxial plug (usually 1/4-inch diameter). An unbalanced phone plug has a tip for the hot signal and
a sleeve for the shield, which connects to ground. A balanced
phone plug has a tip for the signal hot signal, a ring for the return
signal, and a sleeve for the shield.
GLOSSARY
Phono plug A coaxial plug with a central pin for the hot signal
and a ring of pressure-fit tabs for the shield or ground. Also called
RCA plug. Phono plugs are used on TASCAM modular digital
multitrack recorders and on consumer stereo equipment.
Pickup A piezoelectric transducer that converts mechanical vibrations to an electrical signal. Used in acoustic guitars, acoustic
basses, and fiddles. Also, a magnetic transducer in an electric guitar
that converts string vibration to a corresponding electrical signal.
Pinch roller In a tape-recorder transport, the rubber wheel that
pinches or traps the tape between itself and the capstan, so that
the capstan can move the tape.
Ping-ponging See Bouncing tracks.
Pink noise A noise signal containing all frequencies (unless bandlimited), with equal energy per octave. Pink noise is a test signal used for equalizing a sound system to the desired frequency
response, and for testing loudspeakers.
Pitch The subjective lowness or highness of a tone. The pitch of a
tone usually correlates with the fundamental frequency.
Pitch control A control on a tape recorder that varies the tape
speed, thereby varying the pitch of the signal on tape. The pitch
control can be used to match the pitch of prerecorded instruments with that of an instrument to be overdubbed. It is also
used for special effects, such as “chipmunk voices,” and to play
prerecorded tracks slowly so that fast musical passages can be
overdubbed more easily.
Pitch shifter A signal processor that changes the pitch of an instrument without changing its duration.
Playback equalization In analog tape-recorder electronics, fixed
equalization applied to the signal during recording to compensate for certain losses.
Playback head The head in a tape recorder that picks up a prerecorded magnetic signal from the moving tape and converts it to a
585
586
GLOSSARY
corresponding electrical signal. The playback head isn’t the same
as the sel-sync or sync head.
Playlist See Edit decision list.
Plug A male connector that inserts into a jack. Also, short for plug-in.
Plug-in Effects software that you load into your DAW recording
program (called the host). The plug-in becomes part of the host
program and can be called up from within the host. Some manufacturers make plug-in bundles, which are a variety of effects in a
single package.
Podcast A digital recording of a radio broadcast or similar program, usually in MP3 format, accessed on the Internet by an RSS
feed, for downloading to a computer or personal audio player.
Polar pattern The directional pickup pattern of a microphone. A
plot of microphone sensitivity versus angle of sound incidence.
Examples of polar patterns are omnidirectional, bidirectional,
and unidirectional. Subsets of unidirectional are cardioid, supercardioid, and hypercardioid.
Polarity Referring to the positive or negative direction of an electrical, acoustical, or magnetic force. Two identical signals in opposite polarity are 180 degrees out of phase with each other at all
frequencies.
Polyphonic Describing a synthesizer that can play more than one
note at a time (chords).
Pop (1) A thump or little explosion sound heard in a vocalist’s
microphone signal. Pop occurs when the user says words with
“p,” “t,” or “b” so that a turbulent puff of air is forced from the
mouth and strikes the microphone diaphragm. (2) A noise heard
when a mic is plugged into a monitored channel, or when a
switch is flipped.
GLOSSARY
Pop filter A screen placed on a microphone grille that attenuates
or filters out pop disturbances before they strike the microphone
diaphragm. Usually made of open-cell plastic foam or silk, a pop
filter reduces pop and wind noise.
Portable studio A combination recorder and mixer in one portable
chassis.
Post-echo A repetition of a sound, following the original sound,
caused by print-through.
Power amplifier An electronic device that amplifies or increases the
power level fed into it to a level sufficient to drive a loudspeaker.
Power ground A connection to the power company’s earth ground
through the U-shaped hole in a power outlet. In the power cable
of an electronic component with a 3-prong plug, the U-shaped
prong is wired to the component’s chassis. This wire conducts
electricity to power ground if the chassis becomes electrically hot,
preventing shocks. See Safety ground.
Preamplifier (Preamp) In an audio system, the first stage of amplification that boosts a mic-level signal to line level. A preamp is a
stand-alone device or a circuit in a mixer.
Pre-delay Short for pre-reverberation delay. The delay (about 30 to
150 msec) between the arrival of the direct sound and the onset
of reverberation. Usually, the longer the pre-delay, the greater the
perceived room size. Using pre-delay on an instrument’s or vocal’s
reverb helps to clarify the sound.
Pre-echo A repetition of a sound that occurs before the sound
itself, caused by print-through in analog tape.
Pre-fader/Post-fader switch A switch that selects a signal either
ahead of the fader (pre-fader) or following the fader (post-fader).
The level of a pre-fader signal is independent of the fader position; the level of a post-fader signal follows the fader position.
587
588
GLOSSARY
Preproduction Planning in advance what you’re going to do at a
recording session, in terms of track assignments, overdubbing,
studio layout, and microphone selection.
Presence The audible sense that a reproduced instrument is present
in the listening room. Some synonyms are closeness, definition,
and punch. Presence is often created by an equalization boost in the
midrange or upper midrange, and by a high direct-to-reverb ratio.
Pressure zone microphone A boundary microphone constructed
with the microphone diaphragm parallel with, and facing, a
reflective surface.
Preverb A special effect in which the reverberation of a note precedes it, rather than follows it. It turns a snare-drum hit into a
whip sound, like ssSSHHK! Chapter 10, Effects and Signal
Processors, describes how to create it. Also see Reverse echo.
Print To record on tape or disc.
Print-through The transfer of a magnetic signal from one layer of
analog tape to the next on a reel, causing an echo preceding or
following the program.
Production (1) A recording that is enhanced by effects. (2) The supervision of a recording session to create a satisfactory recording. This
involves getting musicians together for the session, making musical suggestions to the musicians to enhance their performance, and
making suggestions to the engineer for sound balance and effects.
Program bus A bus or output that feeds an audio program to a
recorder track.
Program mixer In a mixing console, a mixer formed of inputmodule outputs, combining amplifiers, and program busses.
Pro Tools A popular digital audio editing platform for professional
use. It offers computer multitrack recording, overdubbing, mixing, editing, and a variety of plug-in effects.
GLOSSARY
Proximity effect The bass boost that occurs with a single-D directional microphone when it’s placed a few inches from a sound
source. The closer the microphone, the greater the low-frequency
boost due to proximity effect.
Pulse code modulation See PCM.
Pulse density modulation See PDM.
Punch-In/Out A feature in a multitrack recorder that lets you insert a
recording of a corrected musical part into a previously recorded track
by going into and out of record mode as the tape or disk is rolling.
Pure waveform A waveform of a single frequency; a sine wave. A
pure tone is the perceived sound of such a wave.
Q
In a parametric equalizer, Q is the sharpness of the boost or cut.
A Q of 1 is a broad boost or cut that covers a wide range of frequencies. A Q to 10 is a narrow boost or cut that covers a small
range of frequencies. Q or Quality factor is a filter’s center frequency divided by the bandwidth.
Quarter-track A tape track recorded across one-quarter of the
width of an analog magnetic tape. A quarter-track recorder usually records two stereo programs (one in each direction).
Rack A 19-inch-wide wooden or metal cabinet used to hold audio
equipment.
Radio-frequency interference (RFI) Radio-frequency electromagnetic waves induced in audio cables or equipment, causing various noises in the audio signal.
Random access Referring to a storage medium in which any data
point can be accessed or read almost instantly. Examples are a
hard disk, compact disc, and MiniDisc.
Rarefaction The portion of a sound wave in which molecules are
spread apart, forming a region with lower than normal atmospheric pressure. The opposite of compression.
589
590
GLOSSARY
R-DAT See DAT.
RealAudio A highly compressed audio file format used for streaming audio. Generally, RealAudio has lower fidelity (less treble) than
MP3, but the fidelity depends on modem speed. RealAudio files
(.ra or .rm) are often used as short excerpts or previews of songs.
Real-time recording (1) Recording notes into a sequencer in the
correct tempo, for later playback at the same tempo as recorded.
(2) A recording made direct to lacquer disc or direct to 2-track
without any overdubs or mixdown.
Re-amping Recording a guitar amp that is fed the signal from a
direct-recorded electric guitar track. This technique lets you work on
the amp’s sound during mixdown, rather than during recording.
Recirculation Feeding the output of a delay device back into
its input to create multiple echoes. Also, the control on a delay
device that affects how much delayed signal is recycled to the
input. See Regeneration.
Record To store an event in permanent form. Usually, to store
an audio signal in magnetic form on magnetic tape or disk, or
to store an audio signal in optical form on a CD-R or CD-RW.
Recording is also possible on magneto-optical disk, on MiniDisc,
in RAM, and on memory cards.
Record equalization In analog tape-recorder electronics, equalization applied to the signal during recording to compensate for certain losses.
Record head The head in a tape recorder or hard drive that puts
the audio signal on tape or disk by magnetizing the tape or disk
particles in a pattern corresponding to the audio signal.
Recorder–mixer
one chassis.
A combination multitrack recorder and mixer in
Recording/reproduction chain The series of events and equipment
that are involved in sound recording and playback.
GLOSSARY
Reflected sound Sound waves that reach the listener after being
reflected from one or more surfaces.
Regeneration See Recirculation.
Region In a digital audio editing program, a defined segment of
the audio program. Also called clip or zone.
Release The final portion of a note’s envelope in which the note
falls from its sustain level back to silence.
Release time In a compressor, the time it takes for the gain to
return to normal after the end of a loud passage.
Remix To mix again; to do another mixdown with different console settings or different editing.
Remote recording See On-location recording.
Removable hard drive A hard-disk drive that can be removed and
replaced with another, used in a DAW or hard-drive recorder to
store a program temporarily.
Render To convert one audio format to another in a DAW or
hard-drive recorder. Examples: mix a multitrack recording to a
2-channel WAV, AIFF, rm, or MP3 file; convert a track with realtime effects to a track with embedded effects; convert MIDI tracks
to audio tracks; combine multiple clips into a single track.
Resistance The opposition of a circuit to a flow of direct current.
Resistance is measured in ohms, abbreviated w, and may be calculated by dividing voltage by current.
Resistor
An electronic component that opposes current flow.
Return-to-zero See Memory rewind.
Reverberation Natural reverberation in a room is a series of multiple sound reflections that make the original sound persist and
591
592
GLOSSARY
gradually die away or decay. These reflections tell the ear that you’re
listening in a large or hard-surfaced room. For example, reverberation is the sound you hear just after you shout in an empty gymnasium. A reverb effect simulates the sound of a room—a club,
auditorium, or concert hall—by generating random multiple
echoes that are too numerous and rapid for the ear to resolve. The
timing of the echoes is random, and the echoes increase in number
with time as they decay. An echo is a discrete repetition of a sound;
reverberation is a continuous fade-out of sound.
Reverberation time (RT60) The time it takes for reverberation to
decay to 60 dB below the original steady-state level.
Reverse echo A multiple echo that precedes the sound that caused
it, building up from silence into the original sound. This special
effect is created in a manner similar to preverb.
RFI See Radio frequency interference.
Rhythm tracks The recorded tracks of the rhythm instruments
(guitar, bass, drums, and sometimes keyboards).
Ribbon microphone A dynamic microphone in which the conductor is a long metallic diaphragm (ribbon) suspended in a magnetic field.
Ride gain To turn down the volume of a microphone when the
source gets louder, and turn up the volume when the source gets
quieter in an attempt to reduce dynamic range.
Ringing See Overhang.
RMF Abbreviation for Rich Music Format. A format for a MIDI file
with General MIDI sounds plus custom sounds. Designed to be
played on a Beatnik player.
Room modes See Standing wave.
RT60 See Reverberation time.
GLOSSARY
SACD
See Super Audio CD.
Safety copy A copy of the master tape or CD, to be used if the master is lost or damaged.
Safety ground See Power ground.
Sample (1) To digitally record a short sound event, such as a single
note or a musical phrase, into computer memory. (2) A recording of such an event. (3) A measurement of an analog waveform
that is done several thousand times a second during a PCM A/D
conversion.
Sampling (1) Recording a short sound event into computer memory. The audio signal is converted into digital data representing the
signal waveform, and the data is stored in memory chips, tape, or
disc for later playback. (2) In PCM digital recording, measuring an
analog waveform periodically, several thousand times a second.
Sampling rate In a digital recording, the frequency at which an
analog waveform is sampled or measured. The sampling rate of
CD-quality audio is 44,100 samples per second. The higher the sampling rate, the higher the high-frequency response of the recording.
Saturation Overload of magnetic tape. The point at which a further
increase in magnetizing force doesn’t cause an increase in magnetization of the tape oxide particles. Distortion is the result.
Scene automation A form of console automation in which the
console settings are stored in memory. A “snapshot” or reading of
many of the settings is taken and stored for later recall. In contrast,
dynamic automation continuously follows the fader and knob
moves, and the automation data is usually stored as a MIDI file.
SCMS Abbreviation for Serial Copy Management System. An anticopy scheme in consumer digital audio devices (those with S/PDIF
connectors). SCMS circuits read flags in the data stream that allow
users to make only first-generation copies, not copies of copies.
593
594
GLOSSARY
Scratch vocal A vocal performance that is done simultaneously with
the rhythm instruments so that the musicians can keep their place
in the song and get a feel for the song. Because it contains leakage,
the scratch-vocal recording is usually erased. Then the singer overdubs the vocal part that is to be used in the final recording.
Scrub To manually move an open-reel tape slowly back and forth
across a recorder playback head to locate an edit point. Some digital editing software has an equivalent scrubbing function.
SCSI Abbreviation for Small Computer Systems Interface. A standard spec for high-speed computer input and output. Commonly
used for hard drives, CD-ROM drives, and CD-R recorders.
Supports up to seven devices.
Sensitivity (1) The output of a microphone in volts for a given
input in sound pressure level. (2) The sound pressure level a
loudspeaker produces at one meter when driven with 1 watt of
pink noise. See also Sound pressure level.
Sequence A MIDI data file of musical-performance note parameters, recorded by a sequencer.
Sequencer A device or computer program that records a musical performance done on a MIDI controller (in the form of note
numbers, note on, note off, etc.) into computer memory or hard
disk for later playback. During playback, the sequencer plays synthesizer sound generators or samples.
Session (1) A time period set aside for recording musical instruments, voices, or sound effects. (2) On a CD-R, a lead-in, program
area, and lead-out.
Shelving equalizer An equalizer that applies a constant boost or
cut above or below a certain frequency so that the shape of the
frequency response resembles a shelf.
Shield A conductive enclosure (usually metallic) around one or
more signal conductors, used to keep out electrostatic fields that
cause hum or buzz. A shield in a mic cable is a cylindrical mesh
of fine wires.
GLOSSARY
Shock mount A suspension system that mechanically isolates a
microphone from its stand or boom, preventing the transfer of
mechanical vibrations.
Sibilance In a speech recording, excessive frequency components
in the 3 to 10 kHz range, which are heard as an overemphasis of
“s” and “sh” sounds.
Side-addressed Referring to a microphone whose main axis of
pickup is perpendicular to the side of the microphone. You aim
the side of the mic at the sound source. See also End-addressed.
Signal A varying electrical voltage that represents information, such
as a sound.
Signal path The path a signal takes from input to output in a piece
of audio equipment.
Signal processor A device that is used to alter a signal in a controlled way.
Signal-to-noise ratio Abbreviated as S/N. The ratio in decibels
between signal voltage and noise voltage. An audio component
with a high S/N has little background noise accompanying the
signal; a component with a low S/N is noisy.
Sine wave A wave following the equation y ⫽ sin x, where x is
degrees and y is voltage or sound pressure level. The waveform of a
single frequency. The waveform of a pure tone without harmonics.
Single-D microphone A directional microphone having a single
distance between its front and rear sound entries. Such a microphone has proximity effect.
Single-ended (1) An unbalanced line. (2) A single-ended noise reduction system is one that works only during tape playback (unlike
Dolby or dbx, which work both during recording and playback).
Slap, Slapback An echo following the original sound by about 50
to 150 msec, sometimes with multiple repetitions.
595
596
GLOSSARY
Slate At the beginning of a recording, a recorded announcement of
the name of the tune and its take number. The term is derived
from the slate used in the motion-picture industry to identify the
production and take number being filmed.
SMPTE timecode A modulated 1200 Hz square-wave signal used to
synchronize two or more recorders or MIDI devices. SMPTE uses
the format HH:MM:SS:FF (hours, minutes, seconds, frames) to
specify locations in the recorded program. SMPTE is an abbreviation for the Society of Motion Picture and Television Engineers,
who developed the timecode.
Snake A multipair or multichannel mic cable. Also, a multipair
mic cable attached to a connector junction box.
Snapshot automation See Scene automation.
Soft synth A synthesizer in software form.
Solo On an input module in a mixing console, a switch that lets
you monitor that particular input signal by itself. The switch
routes only that input signal to the monitor system.
Sound Longitudinal vibrations in a medium (such as air) in the
frequency range 20 to 20,000 Hz.
Sound card A circuit card that plugs into a computer and converts
an audio signal into computer data for storage in memory or on
hard disk. The sound card also converts computer data into an
audio signal. A type of PCI audio interface (see Audio interface).
Sound effects Recordings of non-musical sounds—such as a door
slam, gunfire, thunderstorm, car, or telephone—used in dramatic
productions, radio spots and commercials. Not to be confused
with Effects.
Sound module (Sound generator) (1) A synthesizer without a
keyboard, containing several different timbres or voices. These
GLOSSARY
sounds are triggered or played by MIDI signals from a sequencer
program or by a MIDI controller. A sound module can be a standalone device or a circuit on a sound card. (2) An oscillator.
Sound pressure level (SPL) The acoustic pressure of a sound wave,
measured in decibels above the threshold of hearing. The higher
the SPL of a sound, the louder it is. dB SPL ⫽ 20 log(P/Pref), where
P ⫽ the measured acoustic pressure and Pref ⫽ 0.0002 dyne/cm2.
Sound wave The periodic variations in air pressure radiating from
an object vibrating between 20 and 20,000 Hz.
Spaced-pair A stereo microphone technique using two identical
microphones spaced several feet apart horizontally, usually aiming straight ahead toward the sound source.
Spatial processor A signal processor that allows images to be
placed beyond the limits of a stereo pair of speakers, even behind
the listener or toward the sides.
S/PDIF Abbreviation for Sony Philips Digital Interface; IEC 958
Type II. A 2-channel digital signal interface format that uses a 75ohm coaxial cable with RCA connectors, or an optical cable with
TOSLINK connectors. See also AES/EBU.
Speaker
See Loudspeaker.
Spectrum The output level versus frequency of a sound source,
including the fundamental frequency and overtones.
SPL
See Sound pressure level.
Splice To join the ends of two lengths of magnetic tape or leader
tape with tape. Also, a splice is the taped joint between two
lengths of magnetic tape or leader tape.
Splicing block See Editing block.
597
598
GLOSSARY
Split console A console with a separate monitor-mixer section. See
also Input/output (I/O) console.
Splitter A transformer or circuit used to divide a microphone signal
into two or more identical signals to feed different sound systems.
Spot microphone In classical music recording, a close-placed
microphone that is mixed with more distant microphones to add
presence or to improve the balance.
Standing wave An apparently stationary waveform created by multiple reflections between opposite room surfaces. At certain points
along the standing wave, the direct and reflected waves cancel, and
at other points the waves add together or reinforce each other.
Stem A submix in Pro Tools lingo—a drums submix, keys submix,
left-front submix, etc.
Step-time recording Recording notes into a sequencer one at a time
without regard to tempo, for later playback at a normal tempo.
Stereo, Stereophonic An